Vol. XV, Issue 2 (Spring 2008):African Engineering - Terraces and Earthworks


Gloria Emeagwali
Chief Editor

Walton Brown-Foster
Copy Editor

Haines Brown


Olayemi Akinwumi

Zenebworke Bissrat

Paulus Gerdes

Mosebjane Malatsi
(South Africa)

Alfred Zack-Williams
(Sierra Leone)


Tennyson Darko
Asst. Dir. ITS, CCSU

Peter K. LeMaire
Professor, CCSU

Bernice A. LeMaire
Website Designer

Maintained by

For more information concerning AfricaUpdate
Prof. Gloria Emeagwali
CCSU History Dept.
1615 Stanley Street
New Britain, CT 06050
Tel: 860-832-2815



Table of contents


In the winter issue of Africa Update, we examined aspects of Nigeria’s Indigenous Chemistry. We continue the theme of African indigenous science and technology in the present issue. The focus is on engineering. Samuel Gwimbe of the History Department, in the College of Education, Gindiri, Nigeria, provides us with in-depth analysis of extensive agricultural terraces in two regions of Nigeria, namely, Yil Ngas in the Jos Plateau of central Nigeria, and Gwoza in the northeast. The aggregate mileage of terrace lines for the two areas combined could well reach over 50,000 miles. And yet, historians of technology, in particular, and African history in general, have ignored this phenomenal record breaking and historic activity.

 Patrick Darling’s focus is on the Benin earthworks estimated to be about 10,000 miles- recognized in the Guinness Book of Records in 1974, and in some later editions, as being the world’s longest ancient earthworks.

We are delighted and honored to include the work of these two scholars and researchers in the Spring Issue of Africa Update.


Chief Editor
Dr. Gloria Emeagwali

Return to Table of Contents

Building Terraces in Agriculture: A Feature of Yil Ngas before the 20th century
By Samuel Barde Gwimbe
College of Education, Gindiri, Nigeria

The evidence illustrative of the extent to which the Ngas people of the Jos Plateau in central Nigeria took hold of the environment around them, by the end of the nineteenth century, abounds in various ways.

One particular form of that evidence is the technique of building terraces on their farms, epitomizing their knowledge of scientific agriculture and engineering in a pre-capitalist setting. This paper attempts to recapture the nature of Ngas terrace building and the evidence of terraces as a characteristic feature of their environment at the time, as a humble contribution to the reconstruction of the history of pre-20th century science and technology on the African continent- and the conscious attempt to develop a database on it for the 21st century.

Yil Ngas is located on the south-eastern escarpment of the Jos Plateau. It stretches down from the rather rugged edge of the plateau to its foothills, covering an area of about 2,000 square kilometers. It lies between latitudes 9 to 10 North and longitudes 9 to 10 East. The pre-colonial (indigenous) inhabitants called it Yil Ngas, which literally denotes ‘Ngas country,’ or ‘Ngas land’. Until 1906, when British colonial encroachment began, Yil Ngas was largely characterized by independent polities.

Various Ngas traditions of origin and migration often tend to retrace the routes of migration of their ancestors from the Chad Basin to Yil Ngas or from the Chad Basin through Kwararafa in the lower Benue Valley to Yil Ngas. Some lineages claim being autochthonous. Their language belongs in the Chadic branch of the Afro-Asiatic family of languages. By the end of the nineteenth and beginning of the twentieth century, the Ngas lived in several politically non-centralized communities but had culturally been transformed into a cohesive society with a measure of concrete, internal historical development. Several British colonial invaders in the first decade of the twentieth century, described Ngas communities as thickly populated. This paper broaches on their pre-colonial techniques of soil and water conservation, especially the building of terraces in agriculture. The Physiography:

Captain Renny, a British colonial officer, who first led colonial soldiers to commence the conquest of Yil Ngas in 1906, gave this on-the-spot report to his superior military officers:

The Angass tribes inhabit a large tract of rough mountainous country … inside the area inhabited by the tribe, there are practically no roads or even tracks, the whole country being a vast mountain range extremely rough and rocky and intersected in every direction by deep ravines which would render marching impossible for a column during the rains.1
Renny visited more of the outcrop highland territory presently shared between Pankshin, Wokkos, and Ampang-East Districts of today than elsewhere in Yil Ngas. Renny’s description implies that Yil Ngas as an environment was practically uncultivable. That was not, and is not, the case. In such a hilly environment, factors of erosion existed. Its impact on the landscape of Yil Ngas, however, did not demand such exaggerated qualifications. Dimka, a recent observer of the geography of Yil Ngas, said with particular reference to its hills, that: "Most of these hills have been badly eroded by swift rivers giving rise to a very rugged topography with interlocking spurs”.2 Yet cultivation of crops is a continuing reality in the area. Here and there, inselbergs and other rock-masses could be seen today, in Yil Ngas, astride outcrops, hills and hillocks which are in turn either perching atop mountain ranges or scattered on the valleys and the vast downlands extending from the foothills; the very ground itself is strewn with sandstones, gravels, plutons, mica, h'k-per or feldspars, h’k-fifi or granites and so on. This is a blunt testimony to persistent denudation processes. The major forces of denudation in Yil Ngas are water and fire. We focus here mainly on how man contained the eroding power of water through terrace building especially on farms. The hydraulic denudation fostering the erosion of Yil Ngas over the centuries however is not of great magnitude in its carrying content, being that the land has long been a watershed country, often characterized by steep slopes and swift streams. Between the rock-masses, hillocks and hills, sometimes wet-crevices, gorges, ravines, and rivulets, features of a watershed country, are found, whose waters, especially during the wet season wantonly furrow the land-surface and flow into bigger streams and rivers. The Taman drains the western and southern areas into the Shimankar, which in turn empties into the River Benue. The Gyangyang, in conjunction with the Nyinang (Yinang) drain the northern and eastern areas of Yil Ngas, into the Wase, which in turn empties into the River Benue. The Lere stream flows partially out of Yil Ngas northwards until it eventually goes into the Gongola, which in turn, empties into the Benue. All of these waters do eventually drain into the River Benue.

Yil Ngas suffers from an annual dry season shortage of water. By the turn of the century, the present trend of seasonal scarcity of drinkable water already existed in Yil Ngas. It was not peculiar to the communities on the hills; those on the broken foothills were not exempted.13 Because of the slopes, most of the streams usually flood their banks during the rainy or wet-seasons. Great velocities, especially after the torrents, between the months of June and August, usually characterize the streams and rivulets. The case seems to be with some merit, that there were greater water resources in Yil Ngas by the beginning than by the end of the twentieth century. Inter-spaced along the broad dry sandy streambeds of Yil Ngas during the earlier decades of the twentieth century, were several ponds, good for swimming, fishing, drinking (especially by livestock) and so on.3

From the rock-crevices or highland valleys good, cool and clear water flowed out perennially. Such springs still exist. B’utjei-mwa (b’utjei, in the singular) or geysers erupted once in a while and were dreaded because of their destructiveness. Today, some streams have changed their courses, and most riverbeds have not only widened but are deeper with steep banks, in the downlands. In the hills, the streams are swifter and 'valleys' are marked with deep canyons and gorges.

By the end of the nineteenth century, there were some vast plains, largely uncultivated, that have now, today, been subjected to cultivation.4 Most hills and hillsides were however cultivated, in spite of soil erosion by running water. Dimka again observed that:

From the source, the rivers are swift, cutting deep into the mountainsides, creating V-shaped valleys with a few falls. At the foot of the escarpment, much of the gradient is reduced and most of the rivers have characteristics of mature rivers depositing silt over a distance of about 300 - 360m on either side of the banks.5

That kind of transference of silt still continues, so long as the land remains sloppy. By the end of the nineteenth century, there existed already cultivated techniques of containing the process in Yil Ngas otherwise it should have rendered the highlands totally uncultivable, therefore uninhabitable to farmers, by now. Captain Foulkes, a military colonialist and ethnographer, after being in contact with Yil Ngas for about a year, in 1907 took a personal interest in documenting what he called the 'customs of the Angass'. With reference to their agriculture, he noted with astonishment that: "The soil is exceedingly fertile…”6. An astonishing revelation about Yil Ngas in spite of its notorious stoniness, which the travelling spies or 'Survey Party' in 1904, had frighteningly appreciated as a full obstacle, in their own reckoning, to crop-cultivation.7 Foulkes further described the features of Yil Ngas soil using soil types which his fellow British colonial readers were familiar with:

The formation is sandstone and, for the most part, red sandstone, the latter on the high ground showing the rich colouring of Devonshire fields. In the plain Angass the sandstone is not much in evidence, the surface being strewn with granite rocks ……..8

Recently, Dimka has maintained that:

…most soils range from gravely to sandy loam. Laboratory tests have revealed PH6.7. Shallow soils are under lain by hard lateritic pan, some few centimetres below the surface. In rocky hollows, soils are deep, and rich, enriched through weathering of soil nutrients from the rocks. Poorest soils are found at the foot of the escarpment far away the foot of the plateau due to over-cultivation and exposure by soil erosion.9

The rugged physiography of Yil Ngas, its characteristic mountains, hills, streams and soils, therefore, have remained seemingly the same since the beginning of the twentieth century, save for the intensity of soil erosion and the dessication of water resources, leading to the emergence of poorer and poorer soils ‘far away from the foot’ of the Jos Plateau and drier river-beds in the dry seasons.

Pre- Colonial Agriculture and Terraces in Yil Ngas:

Their whole social and religious fabric is founded on an agricultural basis. They are essentially agriculturists and judging from results, admirable ones.

-Foulkes, 1907.10

Agriculture was the principal economic activity practised by Ngas people by the beginning of the twentieth century. Foulkes made a correct assessment of the whole basis of their life in 1907, when he said they were essentially admirable agriculturists. Yet, in the next breadth, blinded by his colonialist mentality which saw no good thing as coming from Yil Ngas, Foulkes scandalized them: “Their [agricultural] system appears too good for their own intelligence to have evolved.” And, if the Ngas did not evolve it, how did they come by it? Foulkes’ opinion was that “they must have been taught it,” and agriculture “being mainly a mechanical process there was little to forget”, after they must have been taught.11 Foulkes did not, however, suggest who might have taught agriculture to the Ngas to make his false assertion seem true.

Like his predecessors the British spies, who three years before, were in Yil Ngas, Foulkes narrated what he saw in 1907 regarding their use of every available space, their intense expenditure of labour and dogged commitment to the building and maintenance of terraces, especially on hilly farm sites: “In the hills [Hill Angas area] every little scrap of ground is utilized, and whole mountain-sides are terraced up at an enormous expenditure of labor. On the steeper slopes the terraces may be 3ft [three feet] high, built up of rocks and stones, and frequently only a few inches wide, just enough room for one row of guinea corn.”12 That shows how meticulous they were, about making terraces in order to get as much of the soil as possible.

Practising agriculture in a geographic environment characterized with the kind of topography that Yil Ngas had and still has, must have, no doubt, been labor intensive. But Foulkes failed to give the credit of knowing how to overcome such a tough environment, through terrace building, to the Ngas people. Had he done so, he would have probably by the same logic undermined any rationale for constraining the Ngas to submit to the British colonial regime, which he represented. Colonial whims and caprices, made it such that the colonialists sometimes portrayed the colonised as warlike and healthy, but at other times, as lazy and unintelligent, though they always exploited their labor. Thus, Foulkes labored to suggest in 1907 that the Ngas were hard-working and had daringly subjected most of their ‘tough’ area to cultivation, yet he also thought they were low in intelligence.

Foulkes did not realise, as it were, that when the going got tough, the Ngas people being tough, got going. They had embraced shifting cultivation, which incidentally he knew nothing about, and developed an attractive method of terrace building on their farms, yet to him they were just pedantic laborers! That was not the case. They approached agriculture creatively. Throughout Yil Ngas, in addition to terrace-building, techniques of shifting cultivation largely blended with swidden practice especially in the outlying farm sector on the one hand, and intensive cultivation largely blended with mixed-cropping especially in the homestead farm sector, on the other. Whether, it was the Ngas communities on the highland hills or those on the broken and hilly plains, they generally cultivated the hilltops and hillsides as well as the plains and valleys, employing the techniques of building and or repairing of terraces and cultivating the beds created as result of terracing.

The comparative size of Ngas settlements vis--vis those of their lowland neighbors by the end of the nineteenth century, to an extent, indicated the large size of Ngas labor-force at the time. Ngas settlements were large, but pitched on mountain ranges, hills, hill-sides and plains. Emerging from Tarokland to the south-east of Yil Ngas in 1904, the Survey Party of British spies comparatively observed that: "From now onwards, instead of scattered hamlets the people usually live in large villages without stockades.”13 Doubtless, the use of stockades, to surround settlement premises, was a feature of the lowlands peoples, which included, among others, the Jukun, Goemai, Yom, Gerka, Tel and Tarok, whose settlements the spies visited before getting into the Ngas country. Capt. Renny, who got to Yil Ngas two years later, reported that Ngas settlements were “very thickly populated and the population of the whole tribe must be at least 250,000”. Furthermore: “All the [Ngas] towns are built in separate compounds (holding 20 to 80 people) scattered as a rule over several square miles in the hills, some of them constructed on the highest peaks which are well-nigh inaccessible.” 14 Let us infer from that, that by the beginning of the twentieth century, the Ngas settlements, said to be large in area, constructed mostly on peaks, with separately built compounds, had on average 50 folks per compound, indicative of a large work force.

From the way their houses were built, settlements were patterned, sites were selected and so forth, the Ngas did not only demonstrate that they understood, and had mastered, their environment to a great extent, but were a people with time-packed experience and principles with which they continued to struggle to conquer nature, and by the turn of the century in question, had already recorded various victories in that regard. The Foulkeses of this world could not successfully deny those victories, to distort their history. One such victory was terrace building on especially rocky hillside farms.

The Ngas developed a system of terracing, which was viewed by the colonialists as one of the best in Africa. 15 Some thought of the utter difficulty of tilling such a stony terrain and simply wondered, how and from where the Ngas borrowed the technique of building terraces. Thus, Boyd Alexander, leader of the Survey Party of spies, observed with reference to what he saw and initially thought in the first week of July1904, that there was very little grass, and few bushes, and it must have been very difficult to grow crops in such a stony soil. 16 Before reporting on how terraces were built in pre-colonial times in Yil Ngas, let us briefly examine terrace building in some parts of the world.

In conducting a survey of four terracing building societies, Critchley and Brommer have suggested that the most ancient terraces are in the foothills of the Himalayas in the State of Uttaranchal in India, which are well over a thousand years old.17 They further state that in the said Himalayan foothills, “the terrace walls or ‘risers’ are sometimes stone faced – when stone is available – but more often they are earth structures”.18 This is exactly what was and is still to an extent practised in Yil Ngas, where, in most cases, stones are available on most farms, except in valleys and on banks of streams where either or both of two earth structure types are usually raised. The smaller earth structure in Yil Ngas is known as gung while the bigger is known as mbanki, more or less a dike. A stone-built terrace in Yil Ngas is known as pang or pang’gang.

Another situation cited by Crithley and Brommer, which in our reckoning is similar to what obtained in Yil Ngas by the end of the nineteenth century, and to an extent still obtains today, is the case of the area around Yogyakarta city in south-central Java, where “using terraces like stepping-stones” farmers cultivated the hillsides. “The result is a landscape of bench terraces that have a slight backslope, allowing excess runoff to drain away.”19 But unlike the Java situation, the antiquity of terrace-building in Yil Ngas should, in our opinion, date earlier than the nineteenth century. At present that remains a subject of a separate research.

Regarding the two African examples of terrace building farming communities cited by Crithley and Brommer, namely, Kabale District in south-west Uganda and Venda area of Limpopo Province of South Africa, the former was stimulated and acquired the practice during the 1940s through colonial ordinances, thus it is of recent origin. In Venda however, “a tradition of building houses and terrace walls with stone has existed for generations”, very reminiscent of what obtained in Yil Ngas in pre-colonial times. “Most of the terraces have stone-faced walls (mitsheto) that have been constructed with pride and considerable masonry skill.”20 Even that is reminiscent of what obtained in Yil Ngas in pre-colonial times, as the next paragraph purports to show.

The Survey Party of British spies, whose initial decision was that it was difficult to cultivate crops in Yil Ngas, did not conceal their deference for the ingenuity of the Ngas people in terracing their entire country, which they marveled at, in 1904:
The whole land was mapped out into little terraces sometimes only a foot or two broad, built to hold up the rain as it ran down the hill, and prevent the soil from being washed away. In places no longer cultivated, only the dilapidated terrace walls remained, through the rents of which in course of time the soil had been blown or washed down. In this case one saw only low walls encircling a hill, and I cannot but think, though with deference... that a similar cause may have led to the erection of the extraordinary systems of concentric walls, which exist in Rhodesia [Zimbabwe]...21

Furthermore, Alexander (1907) compiler and giver of their report, did not conceal his sense of wonder when thanks to terracing, he saw every available space, on their farms was already cultivated: Dugurh [Dungung] stands at the foot of Mt. Ampang [Zwal Pang], which is over 4,000 ft high. We climbed this and found a large plateau at the top covered with populous villages. lt was wonderful to see the way in which every inch of the ground was cultivated. Even spaces not more than a square foot in area, where a little soil had been collected, had been planted with millet or guinea corn, which is the ordinary food of the people throughout the country.22 The difference between the spies’ initial impression and what they later saw, which shocked their imagination, was the employment of appropriate ‘technology’. The technique of terracing was employed at all altitudes, in the foothills and in the hills. Valleys, swamps, plains, hilltops and hillsides, were mostly cultivated, except when they were left to fallow.

Their tools, which were designed to face the hardest grounds, were therefore very strong. They included the chen sak (chen gon) or big hoe, the chen kos or small hoe, the sep or axe, the hammer, the kyeang, the nuk, and the mbamji, a pestle-like digger with a flat-bladed metal head and a wooden handle. These tools were of different sizes, for different services. With their aid, man appeared to have subjected every inch of available ground, as it were, in Yil Ngas to cultivation up to the beginning of the twentieth century, among other achievements. With large communities, which every early calling foreigner saw, at least, it was necessary that adequate, and possibly surplus, food supply was guaranteed, which in such a terrain, apart from enough man-power, required the use of appropriate tools and technical ideas, such as kak kh pang’gang (or kak kh pang), which literally means building or repairing of terraces. Once built, stone faced terraces were repaired from time to time or at every stage of the cultivation process, especially during clearing, weeding, and mulching in a given planting year, as long as there were breaches on them. In order to preserve the neat arrangement of terraces on bush-farms, cattle-herders were sometimes forbidden from grazing their animals on left-over stalks of harvested corn. Earth terraces were also subject to repair either to conserve moisture or prevent flooding of farms by water.

By the beginning of the twentieth century, stones or rock pieces were being used in various other important ways as h’k nchwa’lang or sling stones, h’k char’char or hand-thrown stones, h’k shin’gwak and h’k sar shin, which are grinding or mill stones, h’k t’r’ng, h’k pang-dyer, h’k gh’zj’ok, h’k pii-bwir and so on. Stones were used as h’k dyik lu or building blocks, thus the Ngas, by the turn of century in question, built mostly stonewalled housing. H’k shii-lu, were stones used for constructing the foundations of such dwellings. For granary or corn-bin walls and ceilings, however, refined clay mixed with mud and brittle grasses to produce a type of mortar, was artfully (or ceramically) used as plaster and/or embroidery.

In every big ancestral compound, there was a nefur, an arena or dance theatre, circumscribed by a pang (this is its second meaning:) or a lowly-built circle, or semi circle, of stones, neatly arranged up to two or three feet high and about two feet wide. Hence the term pang nefur is sometimes used to describe a nefur. The frontage of every house where there was no nefur had a po pang which means a smaller place than a nefur where some stones were put for people to sit on, during moments of rest, grief, brief consultations or, joy, especially as spectators, during song and dance presentations.

The raising of bare stone walls without using any mortar or clay, in the historical development of the Ngas people, up to AD 1900, was probably an epitome or a relic of their pre iron-smelting and-working period. Among them, the po pang, was lower than the pang nefur which was, in turn, lower but wider than the bong n’ng, or cattle kraal. A bong n’ng’s height is about the same with that of a lu tap, the small, bare stone-walled thatch roofed unit and watch-shed. A lu tap smacks of how Ngas pre-colonial ancestors simply took advantage of their rocky terrain to meet their shelter and recreational needs. It seems tenable to suppose that in earlier times man’s initial home was a lu tap, an advanced alternative to cave-dwelling.23

Using mortar or clay in buildings must have been developed or acquired later, when perhaps, the Ngas had already attained the iron-working stage and could take advantage of iron-headed tools not only to dig up much clay but also mix mortar. When clay was discovered and used in pre-colonial Yil Ngas is an archeological unknown. But Ngas pottery and other clay articles in about 1900, made of very good clay have survived not only as vessels of an age but unique works of art.

Today, the situation has greatly changed. The large settlements in Yil Ngas by the beginning of the twentieth century are not as big as in antiquity. Places which were more or less large collections of labor-power, which naturally translated into a generally sound scope of production, have dwindled a lot due to emigration. All capable folks (whether males or females, children or the aged, the visually-handicapped or the lame, rulers or the ruled) participated in food production, within the pre-colonial economy of Yil Ngas; today such attitude to work is merely a historical referent. There were practically no idlers, beggars or other forms of lumpen population; today there are.

The inhabitants of Yil Ngas by AD 1900 had fashioned out the means, techniques, and tools, appropriate for sustaining their struggle for existence in a natural environment, especially through building and maintaining terraces on hill, hillside and valley farms.

By the beginning of the twentieth century, most rural neighbourhoods in Yil Ngas where farms were usually operated, had terrace features of being cultivated or having been cultivated before. Explorers and colonialists saw contour lines or dilapidated walls of age-old terraces, apart from the maintained terraces seen in farm-plots still being cultivated. Therefore, by the end of the nineteenth and the beginning of the twentieth century, the problem of Ngas agriculturists was not that of a too difficult, intractable natural environment, as of that of threats to peace which the soon-coming colonial regime constituted. Their environment as maintained and sustained was resourceful enough to meet their various needs of food, shelter, clothing, tools, water, and so on.

Today, terrace building and terrace repairing are still being done, as recent pictures displayed below may show. But most farmers would prefer to cultivate less troublesome locations to reduce the labor demands of terraced locations. It has become fashionable now to farm the valleys and plains and to neglect the hill-tops and sides. Probably this gradual ‘shift’ in preference of the valleys to the hills in cultivation (whereas in the past the hills were well cultivated as well) has had a negative impact on the nature of man's relations with his environment in Yil Ngas, especially as it concerns soil and moisture conservation.
*Department of History, College of Education, Gindiri, PMB 01000, Plateau State, Nigeria

Return to Table of Contents

Benin Earthworks: Perceptions Over Time
Dr. Philip Darling, Akodi Afrika, Nigeria*

Benin-centric history perceived the ancient ditches around its city as being a series of concentric 'moats' or 'walls' dug to protect the city - the innermost of which was dug by the 14th century Oba Oguola, in response to the one dug at Udo, and then deepened up to 20m by Oba Ewuare in the mid-15th century (Egharevba, 1937).

Connah sent out two survey teams to map these 'walls' in the 1960's: each team of six had two men clearing the undergrowth along the bank top, two men holding a chain, one man holding an eight foot high surveying pole, and another man taking readings to the nearest half degree with a prismatic compass. With no way of avoiding thick vegetation patches, and with the chain snagging on the cut undergrowth stalks, each team averaged only 400m a day, with the best day being 800m. Ergonomically, this progress can be expressed as 66m per man day.

Connah failed to triangulate between the radial roads, so the resulting map was badly distorted in places; but it revealed that the walls were a composite of different settlements separated by wide bands of no-man's-lands, which were progressively enclosed in a process of fusion over the centuries. The oral tradition view of three concentric 'walls' was no longer credible; but it persisted amongst Bini historians, many of whom sadly showed no sign of having ever read Connah (archaeologist), Bradbury (ethnographer who, inter alia, revised the Benin chronology) or Ryder (historian).

In the late 1960's/early 1970's, Darling discovered that such earthworks lay over a considerably more extensive area and, on the basis of Connah's earthworks having a density of about 1km of earthwork per 1km2, he estimated that they were about 10,000 miles long - the world's longest ancient earthworks, some four or five times longer than the main Great Wall of China (Guinness Book of Records, 1974). Having competed in UK orienteering at the national level, he had a somewhat different surveying technique, using an orienteering compass to the nearest 10 degrees and pacing it off so as to stop at a ten pace interval (usually 20, 30, 40 or 50 paces). Paper folded into 16 fitted under the compass in his left hand and acted as a notebook: a biro was held in the right hand.

These instrumental traverses using co-ordinates derived from compass and pacing, proceeded at a rate of between four and ten km a day. Initially, a local worker helped to clear the undergrowth, but with experience at taking the line of least resistance through the vegetation (alongside or along bank or ditch), this became no longer necessary. In ergonomic terms, therefore, these surveys were 500 to 1000 times more efficient than Connah's surveying teams, even though they were only about 4 to 8 times faster than the two teams per day.

Accuracy was checked by measuring accumulated error using the sine totals of the ten degree angle sets; and errors were surprisingly low, in the region of 1-3%. Steel tape and prismatic compass surveys along tracks and roads, to pinpoint intersections with the earthworks, as well as the use of Federal Survey maps, aerial photographs and a plan variograph, helped to reduce accumulated error further.

After initial surveys at Ologbo (where the author was teaching) and near Oluku junction, the then Federal Department of Antiquities met expenses for further surveys around Okunmwun, Idogbo, Evbonikhuo, Ugbenu, Uroho and Oka. At the same time the author helped to coach and train with the Midwest State cycling time and he ended up with a medal at the 1973 First All Nigeria Sports Festival. As a result, he was awarded a Nigerian State Scholarship to study what was known of these earthworks in UK based literature, before returning to undertake full-time surveys of them. The Federal Department of Antiquities met the expenses, the State Board of Education provided a salary for an Education Officer on Special Duty, and the Military Governor's Office in Benin provided housing and transport.

Simultaneously serving these three masters, the survey work over the next five months was on course to complete surveys of the whole 6,500 square km earthwork cluster, within the five years allocated for this project. Very soon, it was clear that there were different zones of earthwork density. Numerous small enclosures with several additional loops and narrow cordons sanitaires lay in a core to the north-east and east of Benin. Around this core were peripheries of large enclosures with few additions and broad bands of no-man's land. Later these peripheral zones were recognised as migratory wave fronts, into which migrants were attracted. To the south-west lay the Benin wave front, which from Connah's dates and one from Ohovbe, could be dated to the 12th century or 13th century. To the south was the 14th century Ugha Kingdom wave front. To the north-east lay the mid 15th century Ishan wave front. The earthworks appeared to depict some of the complexities of the wider state-formation process that gave rise to the Benin and Ishan kingdoms.

After two months work, the author attended a conference of Nigeria's archaeologists and with some shock realised that a) he had already done more fieldwork in those two months than all the others had done over the last year b) there was little genuine interest in archaeology or this work - quite understandably, nationalistic politics and suspicion ruled the day. However, all was not lost: the contact encouraged Fred Anozie to request surveys of the Okpe Igala in Anambra State; it resulted in Robert Soper asking the author to join in surveys of the walls of Old Oyo; and a subsequent historical conference brought him into contact with Ade Obayemi, whose real interest was most encouraging.

Five months into the Benin work came the military coup that toppled the Gowon regime. Everyone in receipt of Government funding became the subject of investigation; and the Federal Department of Antiquities was too frightened to over-ride the refusal of a local curator (who was subsequently sacked for corruption) to pay expenses due; and N1,600 of these were met by the Military Governor's Office (the rate of exchange was then close to N1 = 1). With no expenses, and the growing clarity that the new regime wanted little connection with anything approved by the previous regime, the author hung on. Minimising expenditure to that which could be met from his salary, he switched from surveying the earthworks to gathering the oral traditions, sherds and cultural landscapes associated with them.

Progressively, the field vehicle was requisitioned, surveyors came to inspect the house for the next users, and other pressure was applied. However, the author carried on working until his contract expired in 1976. He then decided to write up the results at the Centre of West African Studies at the University of Birmingham, UK. Within a term, his degree was retrospectively upgraded from an MA to a PhD; so he returned to Nigeria to obtain samples for radiocarbon dating and to make a 160km long pottery sherd transect from Benin to Ukpilla, using the recently constructed 132kV (and the 33kV trans Ishan) transmission lines.

The datings suggested a 700 - 1450 A.D. construction period: seriation analyses of the sherds collected showed a movement from north to south, consonant with ramal analyses of the linguistic data (word lists provided by Kay WIlliamson). However, a change in University policy meant that charcoal samples could no longer be measured for free; so only a few were dated (using personal savings) - not enough to demonstrate fully the chronological relationship between the core and the perhipery of the earthwork cluster.

This means that alternative explanations for the earthwork zonations, e.g., the establishment of small settlements in Isi (in the core) are still viable, especially if one is inclined to explain everything in terms of the Benin dynasty. The earthwork pattern of cordons sanitaires (narrow strips of no-man's land between oppposing territories in the core zone - bank, ditch, level strip, ditch, bank) appears to be imitated by bifossate earthworks in the periphery (bank, ditch, level strip, ditch bank), although some are more defensively rational (bank, ditch, bank, ditch). Also, oral traditions link in well to the large earthwork enclosures on the periphery (Okhunwun, Iwu, Utekon. Idogbo, Oka) but often have no link whatsoever with the small primary enclosures of the core zone. Indeed, many of the oldest settlements in the core zone cannot recollect any connection with the high density (7km earthworks per square km) earthworks all around them. Both these observations of the typology and oral traditions point to a core to periphery sequence, predating the Benin dynasty.

The Benin Oba's avbiogbe surveyors from Ife demarcated with trees, not with earthworks; and many of Benin's provincial Enogie officials have residences outside the primary earthwork enclosures. Other major leaders, such as the Osa, have villages occupying the broad bands of no-man's-land, not primary enclosures. However, the Iya n'Akpan - the second earthwork enclosure around the northern, western and south-western part of the City wall is occupied by the Uzama villages on which the dynasty depended; so no simplistic interpretation is secure.

The spiritual significance of the earthworks as the boundary between the real physical world (agbon) and the spirit world (erinbhin), indicates a considerable passage of time since their construction. An old practice, now forgotten and in disuse, echoes an earlier era of exogamy, whereby men stayed put within their village territory, whilst the women could trade (and intermarry) with neighbours; and the women used to sing out as they crossed the boundary earthworks so as not to be mistaken for men, who might be regarded as a war party.

These earthworks, therefore, do more than demarcate territory. They were probably first dug around settlements and their farmlands as barriers against the nocturnal raids of the forest elephant, over a thousand years ago. As more land was required, these linear earth boundaries became useful demarcators of territory and included the vital social function of inter-marriage injunctions within small settlements. As younger sons had no claim to home territory, it seems likely that they were the major component in short distance migrations from the core to the periphery; and these migratory wave fronts became growth areas attracting in migrants from all around.

In 1993, the author tried to organise work to win Nigeria's first UNESCO World Heritage Site for the Benin earthworks, along with the 1000 year old, 160km long, 20m high Sungbo's Eredo kingdom boundary rampart, around the Ijebu heartland. With the collaboration of the National Commission for Museums and Monuments, the authors conducted extra surveys at Udo and Ekhor, helped the Principal Superintendent of Monuments to draw up local conservation agreements, and in 1996 took the opportunity to open up British Foreign Office funding for Benin's Centenary Committee, to undertake a clean up of the main Benin City Wall , and ghost-wrote the nomination document. Sadly, the political suspicions leading to the downfall of Benin a century previously were repeated in respect of the Foreign Office funding and the conservation agreements waiting to be ratified by the Oba. Without, inter alia, any evidence of action and with the non-signing of those agreements, Benin lost its opportunity of becoming Nigeria's first UNESCO World Heritage Site; and its 'World Heritage' archway to the bronzecasters' street is a sad echo of what might have been.

Nonetheless, the Benin earthworks were the author's introduction to archaeology in Nigeria; and the negative politics of Benin merely served to let him cast his net more widely throughout Nigeria, leading to further discoveries of its incredible and fascinating archaeology. The destruction of Nigeria's vulnerable visible archeology was of considerable concern, particularly since virtually no one else was surveying these features. By conducting fieldwork adventure holidays, he brought in UK archaeology students, who then worked with Nigerian colleagues.

Homework inspecting old map, aerial photographs and books, provided a wealth of over 5,000 sites worth examining; and over the last few years much of Nigeria's northern and middlebelt archaeology has begun to come to light. Major Yoruba kingdom boundaries have been mapped at Ijebu, Old Owu and Ilesha, with extensive city walls also surveyed at Olle, Otefon, Kabba, Ikija, Abeokuta and elsewhere.

These large features express the priorities of past ages: they were some of the first to be constructed, they took up a large labour force and they required considerable organisation and planning. By surveying their features and recording their associated rich cultural landscapes, Nigeria's earthworks are serving to re-interpret or confirm much of Nigeria's nearly forgotten history and prehistory in terms of its past priorities. The advent of handheld GPS has encouraged many more Nigerians to venture into the field, although some of the initial efforts have been a little faltering. Owu Ipole and Old Owu maps are guesstimates made from footpath intersections, not actual transects along the course of the walls. The funds and effort expended in a third mapping of Old Oyo's inner wall could have been expended more profitably elsewhere, perhaps in obtaining the actual co-ordinates for the survey of Koso, an earlier Oyo captial. Nonetheless, these belated moves are much welcomed by the author, who will continue to encourage and collaborate with Nigerian colleagues to reveal the true worth of Nigeria's archaeology (which expresses the greatest linguistic and genetic diversity in Africa and the world).

Dr. Darling is associated with Akodi Afrika, of Kwara State, Nigeria, an institution pioneered by the late archaeologist, Professor Ade Obayemi of Kwara State Nigeria. Dr. Darling has been involved in numerous archeological excavations in Nigeria over the last three decades.

Return to Table of Contents

Pre-colonial Terraces on Highland Fringes south of the Chad Basin:
A Comparative Survey of Gwoza and Yil Ngas

Samuel Gwimbe
Gindiri, Plateau State, Nigeria

Location and Traditions
‘Yil Ngas’ is an indigenous place-name meaning Ngasland or territory of the Ngas ethnic nationality, which contains many Ngas towns and villages, and it roughly lies between latitudes 9 and 10 degrees North, and between longitudes 9 and 10 degrees East as could be seen on any appropriate atlas of the area. It is located on the southeast escarpment of the Jos Plateau, stretching down from the hilly and spacious edge of the plateau to its equally spacious foothills, stretching further into the borders of Bauchi State to the north and northeast, and rolling sloppily towards the Benue valley to the southeast. However for the purposes of this paper, we are concerned with the whole of Kanke Local Government Area and Pankshin (North and Central State-Legislative Constituencies of Pankshin) Local Government Area of Plateau State only. Migrant Kanuri, and Hausa-Fulani elements through peaceful means had began to settle in the area, especially in some parts of Kanke Local Government Area, in the nineteenth century.2 Until 1906, when British colonial encroachment into Yil Ngas began, the whole area was largely characterized by independent and politically non-centralized communities, which reportedly were thickly populated.3 It was until 1919, that forces of British colonial encroachment finally took total command of Yil Ngas.

‘Gwoza’ on the other hand, is a place-name referring to the pre-colonial homeland of some twelve smaller but related ethnic nationalities,4 which contains many towns and villages and it roughly lies between latitudes 10 and 11 degrees

North, and between longitudes 13 and 14 degrees East as could be seen on any appropriate atlas of the area. Gwoza is pitched on the southeastern edge of Borno State, hinged on the Cameroun border, occupying what, in some historical works, are known as the Mandara Mountains (or Hills), and their surrounding foothills. Its eleven pre-colonial ethnic nationalities spoke languages in the Laamang-cluster of the Chadic branch of the Afro-Asiatic Family of Languages, though other elements have settled in the area especially in the twentieth century as well. Gwoza is a vast mountainous area with several towns and villages, but the area’s main town is also known as ‘Gwoza.’ In this paper, the term is used with reference to the general area in question, not just its main town. There are fourteen ethnic nationalities inhabiting the Gwoza area presently, which comprise the Dughwede, Zagwada (also known as Gudupe, and Guduf), Mafa, Gavva, Chikide, Zalidivah, Waha, Cinene, Gamargu, Guvuko, Mandara, Glavda, Marghi, and migrant Kanuri, and Hausa-Fulani elements.5 According to an informant, the name ‘Gwoza’ (be that as it may) derives from ‘ngu ze’ a Dughwede expression meaning ‘let’s go to the farm,’ which pronunciation others, in course of time, unjustifiably corrupted to what it is now.6 From 1902 to 1913, when Gwoza fell to the colonial encroachment of the Germans, during which they had militarily invaded the area from the Cameroun side, and finally took possession, and made it part of colonial Cameroon, Gwoza was largely characterized by long-embattled hill-based, reportedly thickly populated communities which had held their turf for centuries, against different shades of invaders and enemies.7 In 1919, however, following the Treaty of Versailles in the aftermath of the First World War, Gwoza changed hands as we say in Nigeria, ceasing to be part of German Cameruon and became part of British Cameroons, included under colonial Nigeria, as a mandated territory.It is amazing to notice that several Gwoza and Ngas traditions of origin virtually point to the same area: the Chad Basin, particularly to Borno, some Gwoza traditions would even add, Ngazargamu.8 At the same time, there are also traditions of origin of lineages that claim some autochthonous historical belonging in the areas or vicinities of the areas they inhabit today.

Thus, it is safer to accommodate the two sets of traditions. This, as could be understood when we remember, as Abdullahi Smith has rightly advanced, that the areas in question, which more or less fall in the Guinea Savanna zone, if not in the Middle Belt, was hitherto a centuries-long buffer zone bringing Chadic and earlier Benue-Congo language communities mainly into peaceful contact, but thereafter sometimes interfusing, assimilating and pressurizing each other, leading either to changes in language (thus of cultural identity) or local migrations within the region.9It is noteworthy that both Gwoza and Yil Ngas are situated somewhat immediately south, and basically outside, of the Chad Basin. Yil Ngas, being part of the Jos Plateau massif and its southeast foothills, is situated on the Chad Basin’s southwest fringes while Gwoza is directly on its south-central fringes.10 The analysis of current historical linguistic evidence actually classifies Shhk Ngas, or the Ngas Language, as well as the main Gwoza languages known as the Laamang group, (excluding in this regard, the Kanuri and the Hausa-Fulani) within the Chadic Branch of the Afro-Asiatic Family of Languages.11 To an extent, this contention which places locations of Gwoza and Yil Ngas in the same bracket as it were, as both lying immediately outside, as well as lying to the south of the Chad Basin, supports, or tends to support, claims that southward migrations of peoples from the Chad Basin must have occurred in the long pre-colonial past. Thanks to unforgotten Chad Basin-pointing traditions of origin, such peoples must have metamorphosed, among others, into the present Ngas and Gwoza peoples, that is, into the communities whose farm terraces are comparatively surveyed here, claims of authochtoneity in some quarters of both areas, notwithstanding. Palmer in the 1920s had advanced the view that the Ngas were cognate with groups such as the Bade, Bole, Ngizim, Ngasar, and Ngijim 12 who are still in the Chad Basin therefore geo-political neighbors of the peoples of Gwoza. For the Gwoza peoples, Connah holds the view that, they migrated from the north down, not from across the Cameroun border immediately to the south up, to their present place.13 Thus a scenic picture of gradual process(es) of southerly (for Gwoza nationalities) and south-westerly (for the Ngas and kith nationalities’) dispersion from the Bornoan mainland and shores, or west-bank of Lake Chad, that is, inside the Chad Basin itself, must have taken place, to areas lying immediately to the south along the fringes of the basin. It is absurd for Robert Netting to consider Ngas traditions of Bornoan origin as somehow, meaningless.14 The fact that Gwoza and Yil Ngas belong together in the Guinea Savanna zone is an important issue; the difference of Gwoza lying in the Northern Guinea Savanna whileYil Ngas lies in the Southern Guinea Savanna belt, or sub-zone, is equally significant.15 Available evidence indicates more or less that, located along the same southern fringes of the Chad Basin are other terrace-building peoples and communities such as the Kulere, Kofyar and Berom of Hoss in Plateau State and Burra of Biu and Higgi of Minchika in Borno and Adamawa States;16 hence our implied, general concern with highland communities lying along the southern fringes of the Chad Basin, and our specific and primary concern with Gwoza and Yil Ngas, both of which the title of this paper betray. It is hoped that in future other terrace-building communities situated along the southern fringes of the Chad Basin, and/or elsewhere, shall further be studied or at least surveyed, to gain a more inclusive picture of this subject in the history of Africa.

Some Terrace Historiography Issues
Simply put, terraces may be defined, as deliberately constructed contour banks, which are effective in safeguarding against erosion on a given farm side, especially on a sloppy landscape. According to Kowal and Kassam, “Experience indicates that well constructed contour banks are an effective safeguard against major erosion and probably the most practical means of controlling run-off and erosion in the Savanna environment.”17 Thus Gwoza and Yil Ngas among other historical communities known to have developed the terracing technique deserve to be applauded, at least for possessing the knowledge and maintaining the usage of a relevant erosion controlling technology in the Guinea Savanna, other than to be derided, as is featured in some literatures. Commenting in the 1950s on contemporary Nigerian communities known for building terraces in agriculture, which were mostly indicative of continuities from the pre-colonial past, Buchanan and Pugh, maintained that: “Skilful and well-developed farms of terrace agriculture are to be found among the pagan peoples …”18 That derogatory category of ‘pagan peoples’ unfortunately included all the communities of Gwoza and Yil Ngas, as well as numerous other communities situated in the Nigerian Middle Belt, or in the wider Guinea Savanna, who practised African indigenous religions and had at the time, neither embraced Islam, spread from the North nor Christianity, spread from the South, to any significant dimensions yet.

As if such negative reportage about the so-called ‘pagan’ peoples, though skilful developers of farm terraces and engineers was not enough, Buchanan and Pugh further suggested that in view of the supposedly sad historical past that the skilful but ‘pagan’ developers of terraces must have had, they had to build the terraces. In their own words, the communities were “invariably associated with a defensive hill-top concentration of settlement, and represent a natural adjustment to the difficulties of an upland environment.”19 A careful reader may notice the critical concepts ‘invariably’ and ‘natural adjustment’ as saying the people acted by either instinct, sheer constraint or conditioning, or weight of necessity, rather than by any sense of ingenuity or innovation, in building and maintaining the terraces! Needless to say that such reconstruction is both false and unfair.

Terraces were built out of a studied concern to conquer and utilize the resources of a given environment to meet man’s desires of food supply. Furthermore, if possession of the technique was a natural adjustment to the challenges of a difficult, upland environment, then it ought to stand in evidence that all mountainous communities naturally possessed the technique of building admirable terraces on their farms, to suit Buchanan and Pugh’s suggestion, but unfortunately that is not the case.20

Neither could we accept the hypothesis that wherever the technique of building terraces obtained in pre-colonial Nigeria or Africa at large, it was invariably and always an indication of a defensive hilltop concentration of settlement, which Buchanan and Pugh purport to suggest. In Gwoza, however, that position, or hypothesis of a defensive hilltop concentration of settlements may seem tenable given the fact that the best examples of terraces in Gwoza today, are up, astride the mountains, and hardly in the foothills, even though intense downhill migration appears to have taken place in the last hundred years. Though covered with some vegetation, the more or less sharp-sandy, stoneless appearance of the soils in the foothills of Gwoza on the one hand, and the practice of cultivating farms without the aid of ridges in Gwoza generally might have functioned to give a sharp contrast between the foothills where there are really no significant terraces, and the mountain-tops and, especially in the south, hillsides. In Gwoza there are well-preserved and visibly well-built dry-stone terraces, thus, betraying the impression of a terrain of hitherto defensive hill-top concentration of settlements. It is dangerously inadequate to build and advocate a general theory based on the experience and evidence of Gwoza as a homogenous community, alone, and superimpose it on the explanation of a host of other terrace building communities in Nigeria, such as Buchanan and Pugh have willy-nilly done. Such reductionism is untenable. In Yil Ngas, as a matter of fact, evidences of terraces are visible in both the high, mountain-based and in the foothill and downland communities, built with equal ability and circumspection. Differences between rates of downhill and uphill migration in Yil Ngas in the last hundred years, might have virtually been very small, given the fact that greater commercial and urban developments, especially with the choice and development of Pankshin town since 1909 as a regional headquarters, have occurred in the more hilly edge of the Jos Plateau, where Pankshin town is situated, than in the foothills; unlike the case in Gwoza where its major urban and commercial developments in the last hundred years have been in the foothills, where Gwoza town is situated. Furthermore, it is difficult to postulate any differences between hill and plains Ngas settlements to the extent of suggesting that either area was characteristic of a foothill or hilltop ‘concentration of settlement’ by the beginning of the twentieth century. If anything, early European travelers simply stated that Ngas settlements, when compared with those of their neighbors in the Benue valley, were large, populated and unstockaded.21 For Gwoza, however, it has been suggested that in past centuries their mainly mountain-based settlements, on the Gwoza peninsula, were characterized by high population densities; further substantiating the tenability of the hypothesis of a defensive hilltop concentration of settlement, in that regard.22

The salient impression, attributable to Buchanan and Pugh, which ultimately suggests that hill-dwelling communities happened to have built terraces due to the practice of a defense economy (being as it were surrounded or kept under siege by their invading enemies!) rather than out of sheer concern to conquer the environment and ensure food production, also leaves something to be desired. …….23

On a general note, it would be difficult to deny that the quest for security and defense, though varying in strategy and situation, has been the vogue of all societies whether on the hills or plains, hence communities of Gwoza could not be an exception.25

The foregoing impression of air-tight separation between erstwhile inhabitants of Gwoza peninsula and their neighbors in the surrounding plains up to 1919 seems exaggerated. There must have been moments of conflict no doubt, but moments of peace and great diplomacy between Gwoza communities and their neighbors on the plains both far and near, could not be totally gainsaid. The old belief that due to invasions, entire populations in Africa were constraint to either emigrate or, as it were, to totally isolate themselves from others, is being held with scepticism in some quarters nowadays.26 If anything, the practice of a defense economy must have been complemented with efforts at diplomacy. Be that as it may. Suffice it simply to recognize the ingenuity of inhabitants in hilly terrain in managing their environments, especially in building terraces in agriculture, appreciating their meaningful contribution, which ironically Buchanan and Pugh have also tried to do (as shown in text of endnote 30 below) but not without contempt, as shown in the foregoing. In the same article of 1941, Stanhope White, reported that the then contemporary inhabitants of Gwoza, settlers as they were, did not perceive themselves as descendants of the forebears who constructed the dry-stone terraces on the peninsula, but simply maintained and utilized what they found: “The present inhabitants all say that their forebears found the present system of terraces on entering the area, and it is obvious that it must have taken centuries of work to bring the system to its present state of perfection.”27 As a result, it is difficult to determine how old the terrace-building practice or system in Gwoza is, until some archeological investigation is done.28 Stanhope White reports further that, during his research, he found a late Paleolithic or early Neolithic artifact, whose details he did not disclose, on one of the spurs of the Gwoza Mountains, which, let us think it together, is quite suggestive of the antiquity of man on the peninsula.29 On the other hand, and to the best of the present author’s knowledge, the Ngas have never attributed the development of their terrace-building practice to forebears of unknown identity other than to their own ancestors.

Given the supposedly ancient antiquity of the Gwoza terraces, it is possible that, before the emigration of various ancestral lineages that later metamorphosed into the present Ngas and other related groups from the Chad Basin, they had known, practised and maintained the terrace-building technology. Pushing far this hypothesis would not be safe since it seems clear that the present Ngas themselves are partly descendants of emigrants from the Chad Basin and partly descendants of authochtonous lineages, who possess only the memory of having been in Yil Ngas for a long time. Conducting a wider field research alongside archeological investigation might engender antidotes to the pending questions regarding the dates or antiquity of terrace-building technologies in either Gwoza or Yil Ngas. For the time-being, we would be content to state that, both are quite old, probably ancient, not later than the first half of the second millennium AD.

Comparable Propositions
One great advantage of this comparative survey is the opportunity granted to not only look at two distinct scenarios through the researcher’s eyes but also to bring into critical limelight the comparable analyses and propositions of several scholars and authors, written often exclusively about either Gwoza or Yil Ngas, often desirous of stating that either the Gwoza or Ngas scenario which they know, represents the best example of terraces in Africa at any bargain. Describing the terraces of Gwoza in specific terms, Buchanan and Pugh assert that:

It is a tribute to the ingenuity and labour of the pagan farmer that the rocky hill-sides and grudging soils of the area were able to support a population whose density, as at the beginning of British occupation in 1919, has been estimated at over 130 per square mile…. Except where topographic conditions make it quite impossible, the hill slopes are girdled with an elaborate system of drystone terraces. …On gentler slopes less elaborate systems of erosion control are used, notably lines of fascines held in place by boulders or small mounds of earth thrown up by hoes; this latter represents a rudimentary form of contour banking.30

Let us compare that with the following reference on terraces in Yil Ngas, which Elizabeth Isichei wrote in 1982, and in the process, quoted Captain Foulkes who lived in the area as a colonial administrator between 1908 and 1911:

the peoples of the area showed a very high degree of mastery of its contrasting environments… the precipitous slopes of the southern escarpment; the homeland of the Hill Ngas, where generations have contained the soil by elaborate dry-stone walling… One early visitor to Ngasland claimed that the people’s agricultural techniques were so elaborate that they must have learnt them from elsewhere! …

All early European visitors were profoundly impressed by the great skill of the area’s agriculturalists…Other observers commented on the Ngas dry-stone walling, the work of generations, which contained the thin soil, prevented erosion and cleared the stones from the rocky fields. ‘…whole mountain sides are terraced up at enormous expenditure of labour. On the steeper slopes the terraces may be 3ft high, built of rocks and stones, and frequently only a few inches wide, just room for one row of guinea corn.’31

In the same article written in 1941, Stanhope White (on whom Buchanan and Pugh heavily depend as far as the terraces in Gwoza were concerned) maintained that:

It is a remarkable sight to see the miles of terraces in the south of the area, following every valley and spur for miles and extending vertically for some 2,500 feet. Tops of the hills have been surrounded by the top-most wall of the terraces, and the space then infilled to give flat expanses of varying size. Where the slope is gentle, the terraces are very wide, but the vast majority of the order of 4 feet high, 5 to 10 feet wide; some terraces, however, are simply a few inches wide, and allow one line of corn only to be sown thereon… These terrace-walls have been carefully built up after the fashion of the dry-stone walls of northern England, and the work is in no way inferior to that of the latter.32

Let us compare White’s views above, with those of Boyd Alexander and his colleagues who visited Yil Ngas in 1904, as a Survey Party of The British Geographic Society, expressing similar astonishment with Ngas terraces and agriculture:

It was wonderful to see the way in which every inch of ground was cultivated…. The whole land was mapped out into little terraces sometimes only a foot or two broad, built up to the rain as it ran down the hill and prevent the soil from being washed away. In places no longer cultivated, only the dilapidated terrace walls remained, through the rents of which in course of time the soil had been blown or washed down. In this case one saw only low walls encircling a hill, and I cannot but think, though with deference… that a similar cause may have led to the erection of the extraordinary systems of concentric walls… in Rhodesia [Zimbabwe].33

In January 2007, the following reference on agriculture in Gwoza, indicating the difference between the hill-based terrace farming and plains farming, was made:

The people of Gwoza are predominantly subsistence farmers, composed of the plain land/terrace farmers. The mountain hill dwellers practice terrace farming which involves a lot of skills in soil management techniques. Due to the limited availability of vast plain land to plant crops, the few which are between the terraces are traditionally managed by supplementing the planting of different crops every two years.34

Then let us compare that position on Gwoza with, though working on the Kofyar, congeners of the Ngas as he calls them, Robert McNettings’ noticeable comment for the terraces in Yil Ngas, in the process of which he cited a statement of Fairbairn written in 1943:

The example of practical engineering in building of terraces and the adaptation of different ridging techniques to varying conditions is striking. One expert [Fairbairn] after describing the more modest terraces and ridges of the neighboring Angas called them ‘a perfect demonstration of theoretical anti-erosion measures carried out on a wide and successful scale over a period of centuries.’35

There is no gainsaying the fact that the claim that the Ngas were taught the technique of terracing by foreigners is not tenable, as shown elsewhere.36 What could be conceded thus far as a possibility in the cosmogonies of Gwoza and Yil Nags, is that terrace-building technology seems to be a relic of a pre-Kanuri Chadic civilization that flourished in the Bornoan area, in which hypothetical case, alongside the ancient constructors of Gwoza terraces in the southern uplands of Borno, the Ngas ancestors possessed the knowledge and the practice before migrating to their present place of abode on the Jos Plateau. On the other hand, however, the technology could also have been developed when the Chad Basin emigrants came into contact with the authochtons in Yil Ngas. It remains a pointer that the technology has remained indigenous to Yil Ngas. But for the peoples of Gwoza, as White reported in 1941, the technology might be very old; so old as to have been abandoned by its original developers, given the waves of human displacement and relocation drifting away from the Chad Basin to its immediate north later found and acquired; which partially explains why it has not been significantly translated, if at all, to suit the plains cultivation in the foothills of Gwoza peninsula. Until wider research has been done, these positions could be only an educated guess, however.

Field Report:
Following a field trip to Gwoza in October 2006, a location not visited by the present researcher before, and reckoning with his knowledge of Yil Ngas, which, being his homeland, the researcher had visited several times, a brief comparative eye-witness account, with few references, is attempted here. The field report is presented in the format of similarities and dissimilarities; hoping that by means of that format, readers would gain a deeper understanding of the issues broached. We proceed to the similarities and dissimilarities, upholding each consideration, simultaneously.

1. Dry Stone Terraces
The appearance, size, length, height and general structure of dry-stone terraces in Gwoza and Yil Ngas are very similar. The researcher could not see any differences. Pictures taken from Gwoza and Yil Ngas, for instance, if not indicated have striking resemblances that could confuse anyone. When the researcher showed pictures taken in Yil Ngas at Gwoza, informants and/or contacts began to make educated guesses as to where in Gwoza those pictures were supposedly taken, but were visibly stunned when they were told the pictures were actually taken at Yil Ngas in far away Plateau State.

The terrace-stone types, with diameters of 3 inches to 1 foot, on the terraces in Gwoza however, at locations high on the top of the mountains, are mostly brown and polished granites, and a few quartz, smooth as though packed from the dry bed of a mighty river; whereas the terrace-stone types on the terraces in Yil Ngas, of similar diameters, are mostly greyish granites. Interlaced here and there, however, though in very rare supply on Ngas farms, are white, brown, and reddish feldspars, and are hardly smooth. Smooth feldspars in Yil Ngas are mostly found on dry-stream beds, not on the mountains. To the present researcher, the evidence of smooth granites astride the Gwoza Mountains is nothing but a blatant evidence supporting the view that, in past millennia, the shores of the Mega Chad, had actually been at the Gwoza peninsula.37

2. Earth Terraces
Since pre-colonial times, both the dry stone-laid and the raised-earth terraces have been, and still are, constructed in Gwoza and Yil Ngas today. Earth terraces, in particular, are being built and presently maintained with equal circumspection in both areas, intermittently constructed alongside the dry-stone terraces on most farm sides, depending on the absence or otherwise of stones on such farm sides, serving the same purpose of conserving water and soil nutrients from easily running down the slopes. In both Gwoza and Yil Ngas, there are low and high terraces depending on the gradients on a given farm side. Oftentimes, high terraces are of dry-stone formation, while low terraces are of raised-earth formation, that is, especially in Yil Ngas. In Gwoza, on the other hand, low terraces, though usually covered with earth on the surface, like all typical earth terraces, are underneath, laid with “lines of fascines held in place by boulders or small mounds of earth thrown up by hoes; this latter represents a rudimentary form of contour banking.38

Sharing his experience and sudden amazement in Gwoza in 1992, Keith Hess maintains that:

It may be a bit of an exaggeration but if sustainable agriculture is your goal, their way of farming rates high. Even around the base of the mountain, there is a hint of something different. Stalks, branches and stones form contour lines across some of the fields. Though an effective method of soil and water conservation, this technique is little practiced elsewhere in Nigeria. But it is in ascending the stair-step mountain trails that one really becomes impressed. All of a sudden there are stone-walled terraces everywhere!39

The sudden appearance of dry-stone and earth terraces at the top of the Gwoza Mountains as well as the use of fascines on low terraces, also astonished the present researcher, because he was seeing fascine-underlaid earth terraces for the first time. Earth terraces in Gwoza are under-laid with dry stalks of sorghum or millet, sometimes with ‘wait-a-bit; thorn-bush sticks, which are sometimes tied together: these are the fascines. These are used especially where there are no stones due, supposedly, to the sandy nature of their soils. Over the stalks and sticks some earth or soil was put as a covering on the top-side of terraces, but any observer could see juttings, if not complete features, of the fascines as it were, on the step-side of the terraces. If, as Buchanan and Pugh suggest, this fascines-using practice in Gwoza is rudimentary, then the distinctly different practice in Yil Ngas, which involves the raising of pure-earth contour banks, or terraces, may be an indication of attainment of a more advanced agricultural level. Be that is it may. Earth terraces in Yil Ngas can hold back water without the aid of such stalks due partly to the loamy nature of their type of soil, and the use of other technologies elicited in number 5 below.

Due to the usefulness of the wait-a-bit thornbushes in Gwoza, to use Stanhope White’s name for it again, they are neither cut nor cleared out of the farm premises, but left alone or preserved for any necessary use, which indicates another sharp contrast with the practice on farms in Yil Ngas, where all shrubs are cleared out especially thorns and thornbushes. Extracted branches of thornbushes in Gwoza are often at places spread across beds of low crops and legumes such as groundnuts, beans, bambara nuts, aya (as called in Hausa) and so on, to keep-and-prick off the monkeys and baboons, which the presenter researcher was told are in good supply, from ravaging them when they are ripe before harvest.40 On the other hand, it does not seem wrong to assert that due to hunting, deforestation, unsanctioned bush burning and other negating activities, such as swidden agriculture, the ape population in Yil Ngas has dwindled, at any rate, to a low point, at least, in the last fifty years. Even so, as hinted elsewhere, the Ngas often utilize the age-old facility of the lu tap, the dry-stone walled thatch roofed hut, permanently constructed and maintained on their bush farm premises, as a watch-shed to stay, sometimes spend the nights, in order to drive off ape and bird pests, while the crops ripen before harvest.41 The dry-stone walls of a lu tap often betray a hard-not-to-notice symmetry with the surrounding dry-stone terrace-walls on a given farm side.

3. Length of Terrace Contours
In Gwoza as well as Yil Ngas, terrace walls or lines are laid in such a manner, which allows them to connect, interlap, and sometimes criss-cross, and are hardly unraveled or constructed in a continuous manner of parallel contours, for more than twenty meters at a stretch. In as much as Stanhope White roughly guessed the aggregate or total mileage of all the terrace lines or contours, if joined one to another in one unbroken continuum, to be in the order of 20,000 miles in Gwoza,42 he did not wish readers to think, as it could easily be done, that there were any single terraces of that length and dimension in Gwoza! In Yil Ngas. Terraces were built not only on the edges and foothills of the Jos Plateau massif, that is, in what colonialists called Hill Angas, but also in the downlands, a wide area stretching away from the foothills, which the colonialists called Plains Angas. Hence the aggregate mileage of the terrace lines or contours in Yil Ngas would be far higher probably in the order of 40,000 miles, given the fact that the whole pre-colonial area of Yil Ngas is spread into four local government areas today, including the indigenous, though smaller, Ngas communities in Tafawa Balewa and Bogoro Local Government Areas of Bauchi State, not part of the area of study in this paper. Terrace agriculture was and is still practised by the Ngas in the entire area. As the 1904 Survey Party, mentioned above, the whole land was mapped out into terraces, this was quite objective; only sacred plots and shrines were exempted. The case in Gwoza, is quite unlike that of Yil Ngas: evidences of terraces and/or terrace farming were and still are relegated to the foothills and hilltop premises of the Gwoza peninsula, there are none on the plain farms. The terrace beds in Gwoza, especially where the gradient is gentler are wider, indeed are bound to be wider, than those in Yil Ngas, for reasons treated in number five below

4. Spread of Terraced Landscapes
The best known terraces in Gwoza are astride the hills, where the settlements are of older antiquity not on the plains nor in the foothills, hence a curious researcher must climb up to the top of the hills. The present researcher had an arduous several hours climb from Plain Guduf to Hill Guduf to examine some terraces and take pictures. Very good-looking (whether stone-laid or earth) terraces in Yil Ngas, on the other hand, can virtually be found everywhere, whether on the hills or on the plains. A brief look into the pre-colonial past of Gwoza when, at some times, they had to contend with military incusions and Jihadists and other invaders for centuries may, to an extent, help us understand why the evidence of terraces lies mainly on the hills not in their foothills. The Ngas, on the other hand, managed to defend and retain comparatively greater political freedom from external encroachment and containment; and thus tilled the hilltops, hillsides, foothills of their land, both far and near with ingenuity, for a long time.

The Gwoza terrain exhibits a sharp contrast between the hills, that is, the Gwoza Hills, jutting up into the air, whose escarpments are mostly rocky except around Limankara area, or entire southern parts, where the hillsides are gentler and more cultivable being less rocky than at other places, and the surrounding plains. The surrounding foothills of Gwoza are virtually plains and more or less stand at the same level all around, especially with the firki fadamas,43 lying to the north towards Bama. Yil Ngas, on the other hand, exhibits the roving intercourse between the edge of the Jos Plateau, characterized by many hills, and their foothills, also characterized by many hills or hillocks. Hence some scholars have described the area as a geographical shatterbelt.44 The whole Yil Ngas area whether on the Jos Plateau edge or in the downlands is characterized with rolling hills, rigdes, and scattered hillocks, valleys, and similar formations. The main point here is the fact of differences in political experiences, which functioned to either hinder the building of terraces on the plains, as was probably the case in Gwoza, or permit the free spread and wide development of the technique, as was basically the case in Yil Ngas.

5. Toolkit of Terrace-Farmers
The implements of peasant farm-labour in Gwoza exclude the big hoe. In Yil Ngas, the big-hoe is very necessary for tilling the ground. Terraced beds in Gwoza are without any ridges. Coming into contact with these realities were surprising discoveries for the researcher in whose area of origin, Yil Ngas, not only are Ngas-made large spade-blade hoes loaded with a lot of iron material making the blades usually quite heavy,45 partly to easily pierce their stony farm grounds, and partly, and probably unconsciously, to bear testimony to the great availability of iron ore in the area. The chenkom was a very large big hoe blade made and used as a form of currency in pre-colonial Yil Ngas, indicative of the great significance of the big hoe in their political economy. Being on the edge of the Chadian firki, and actually practising firki farming on their plains, Gwoza farmers, on the other hand, appear to have contented themselves over the centuries with tool-types common to the firki environment, tools more or less of northern origin, bearing reminiscences of ancient Sahara dwellers. To follow Graham Connah, “It is not surprising therefore that the tool-kit of the earliest settlers on the Chadian firki…include the implements typical of the Saharan Neolithic namely, polished stone axes and others which reflect the lacustrine environment namely plain bone points and bone harpoons.”46 It actually seems that the present tool-kit of the Gwoza peasant farmer, which basically include only the small hoe, ax, knife and a digging stick (which may have or not have an iron-head), is an improved pedigree of the one Connah mentions. On the other hand the Ngas peasant farmer has at his or her disposal the big hoe, the small hoe, the kyeang (designed to be used both as a big hoe and a small-hoe), the ax and an iron-headed digging stick. In the absence of the big hoe, it would be an uphill task to make ridges, but they are not made, on the terraced beds of Gwoza.

As already stated, ridges are not made in Gwoza before cultivation of crops is done, regardless of whether it is on the hill-top or hillside terraced beds or on the plain farms. As a result their terrace beds are generally wider than those in Yil Ngas where ridges are made on the terraced beds. Through the technique of ridging, as a matter of fact, smaller earth terraces are sometimes made within the area of wider dry-stone terraces. According to McNetting “Terracing alone is not the final solution to problems of conservation and water control… A characteristic African solution to the problem of keeping crops above the standing water resulting from heavy rains is to create ridges and mounds.”47 Located in the southern Guinea Savanna belt, distinct from Gwoza, which is in the northern Guinea Savanna belt, in Yil Ngas, farmers had long ago gone the extra mile to practice ridging, due probably to its heavier rainfall. Other problems seem elicited by that position. It does not follow, for instance, that other communities in the northern Guinea Savanna, like Gwoza, do not practice ridging. Neither does it follow that all communities in the southern Guines Savanna, like Yil Ngas, practice ridging.

The Taroh, immediate neighbours of the Ngas to the south, and main inhabitants of Langtang North and Langtang South Local Government Areas of Plateau State, for instance, are somehow like the peoples of Gwoza, who do not have the big hoe nor make ridges before planting their cereal crops. It is usually during mulching that the Taroh used their small hoes and built ridges around their sorghum or millet crops, which they had planted in straight rows in the first place, on plain cleared grounds. In their hilly areas the Taroh make simple ridges and gather stones to control erosion, but not as elaborate as the terraces in Gwoza or neighboring Yil Ngas.48 The soils in Langtang, being on the fringes of the Benue valley, and Gwoza are also very similar: sharp-sandy and soft. While the Laamang languages in Gwoza are Chadic, the Taroh language is Benue-Congo.49 Thus there is no obvious evidence of old congenerous relationship between the two communities, as far as historical linguistics is concerned. Therefore we are virtually left with one of two options, to resolve the dilemma; first, that Gwoza and Langtang were not quite stimulated to develop or adopt the big hoe and/or make ridges due to the softness of their grounds, and, second, that due to lack of iron ore to make the big hoe, they had to make do with what they could afford. The present author prefers the second option. Be that as it may. But apart from the mentioned differences and obvious similarities between the manufactured toolkit of the terrace-farmers of Yil Ngas and Gwoza, which is at issue, the major tool in building and repairing of dry-stone terraces remains the two hands of peasant homo sapiens. Peasant farmers in both Gwoza and Yil Ngas have often made sure that the terraces are well kept to leave the grounds or beds clear for crop cultivation.

Terraces in Gwoza and Yil Ngas are both of great antiquity. It is difficult to state the ages in which the terrace-building technology was first developed either in Gwoza or Yil Ngas. The Gwoza communities, who reportedly once denied having any knowledge of or connections with the original builders of their terraces albeit they have owned, kept, maintained and used them well enough, are now apparently revisiting that initial position.50 The Ngas perceived themselves as direct descendants of the original builders of their terraces. They too have kept and maintained their terraces well enough. The probable uniting explanation considered here is that terrace-building technologies in Gwoza and Yil Ngas are, or might have been, relics of a pre-Kanuri Chadic civilization in the Borno area, from which some elements must have migrated to the present places under review in this paper.

It is now a non- issue to state whether the dry-stone terraces in Gwoza are better built than those in Yil Ngas or the other way round. The structure, neatness, sizes and purposes of the terraces are basically the same. Until wider surveys are done, we could only maintain, for the time-being, that the dry-stone terraces in Gwoza and Yil Ngas are virtually among the oldest and best in Africa. In the area of earth terraces, however, the practice in Yil Ngas, who utilize the big hoe and the ridging technique to build different assortments of earth terraces, is probably more advanced than the one in Gwoza, whose fascines-practice of making earth terraces, has been described as rudimentary. Yil Ngas whose population (as shown in endnote 4) was likely to be higher than that of Gwoza, might also have evinced a higher aggregate mileage of terraces; moreover, unlike the practice at Gwoza, the whole pre-colonial landscape of Yil Ngas was and still is largely marked by terrace contours everywhere. The toolkit utilized in Yil Ngas also appears to have been wider than that of Gwoza. Hence while the origins and antiquity of terrace-building technologies in Gwoza and Yil Ngas may trace back to the pre-Kanuri era in the Chad Basin, it may be correctly stated that, in view of the variables considered in this paper, the level of the terrace practice appears to have advanced slightly higher in Yil Ngas than in Gwoza. Evenso, the practice in Gwoza, in its own right, is basically of excellent standard. This comparative survey, among other things, has valuably helped us to carefully examine the two systems of terracing and to subsequently present the basic conclusion of our findings: they are both of excellent standard.

Return to Table of Contents

1 L. F. Renny (Captain), in: SNP 6, C140/1907, Bauchi Province – Patrols and Expeditions (Confidential), National Archives, Kaduna.
2 Adini S. Dimka, ’ The Geography of Pankshin’ in his: The Impact of Universal Primary Education Rural Farm Labour and the Economy of Pankshin Local Government Area of Plateau State, 1983, BA Dissertation, Department of Geography, Ahmadu Bello University, Zaria, 11.
3 Samuel Barde Gwimbe, Roots of Rural Poverty in Yil Ngas of Plateau State: A Political Economy of Colonial Exploitation and Social Deprivation, 2001, MA Thesis, Department of History, Ahmadu Bello University, Zaria., 43.
4 Gwimbe, Roots of Rural Poverty in Yil Ngas of Plateau State, 44.
5 Dimka, ’ The Geography of Pankshin’ in his: The Impact of Universal Primary Education Rural Farm Labour and the Economy of Pankshin Local Government Area of Plateau State, 11.
6 H. D. Foulkes (Captain), Some Preliminary Notes on the Angass, 1907, SNP 379/1910, National Archives, Kaduna.
7 Boyd Alexander, From the Niger to the Nile, Volume 1, London, Edward Arnold, 1905, 11 .
8 Foulkes (Captain), Some Preliminary Notes on the Angass, 1907.
9 Dimka, ‘The Geography of Pankshin’ in his: The Impact of Universal Primary Education Rural Farm Labour and the Economy of Pankshin Local Government Area of Plateau State, 11.
10 Foulkes (Captain), Some Preliminary Notes on the Angass,
11 Foulkes (Captain), Some Preliminary Notes on the Angass,
12 Foulkes (Captain), Some Preliminary Notes on the Angass,
13 Alexander, From the Niger to the Nile, Volume 1, 95 .
14 Renny (Captain), in: SNP 6, C140/1907, Bauchi Province – Patrols and Expeditions (Confidential)
15 Foulkes (Captain), Some Preliminary Notes on the Angass,
16 Alexander, From the Niger to the Nile, Volume 1, 11 .
17 William Critchley & Marit Brommer, “Understanding Traditional Terracing,” LEISA, (December 2003), 14
18 Critchley & Brommer, “Understanding Traditional Terracing,” LEISA, 14
19 Critchley & Brommer, “Understanding Traditional Terracing,” LEISA, 14
20 Critchley & Brommer, “Understanding Traditional Terracing,” LEISA, 14
21 Alexander, From the Niger to the Nile, Volume 1, 97
22 Alexander, From the Niger to the Nile, Volume 1, 97
23 Gwimbe, Roots of Rural Poverty in Yil Ngas of Plateau State, 61.

2 Samuel Barde Gwimbe, Roots of Rural Poverty in Yil Ngas of Plateau State, AD 1902-1952: A Political Economy of Colonial Exploitation and Social Deprivation, 2001, MA Thesis, Ahmadu Bello University, Zaria, 109
3 Gwimbe, Roots of Rural Poverty in Yil Ngas of Plateau State, 62
4 Stanhope White, ‘Agricultural Economy of the Hill Pagans of Dikwa Emirate, Cameroons, (British Mandate),’ Farm and Forest, Vol V, December 1944 (Reprinted from: The Empire Journal of Experimental Agriculture, Vol. IX No 35 – January 1941), 130. At the time, White found only eight ethnic nationalities, which he calls clans, including, with their estimated populations: Wakara (roughly 12, 900 strong), Hidkala (7,200), Azgavana (9, 400), Kuvoko (1,700), Matakum (1,600), Buhe (9,200), Chikkide (4,600), and Glebda (5,700). A total of 58,100 persons therefore stands as the projected population of Gwoza in the early 1940s, which as shown in White’s paper, might have grown in density to 170 per square mile, higher than the suggested 130 per square mile,during the late nineteenth century. On the other hand, the poplation of the Ngas, was estimated in 1919, to be 76,089, which, if projected to have increased in the early 1940s, it would be somewhere in the upwards of 100,000. For details of that 1919 estimate, See: F B Gall, Gazetteer of Bauchi Province, London, Waterlow & Sons, 1920, 18. Yil Ngas at the time was placed under Bauchi Province.
5 This implies that Kanuri and Hausa-Fulani elements moved into the Gwoza area only in the twentieth century, on the one hand, and that White’s 1941-list is lacking in certain details, on the other. However some peoples often called ethnic nationalities may just be dialectal groups forming parts of bigger ethnic groups.
6 Nathan Habila, 41, Civil Servant: Local Government Works Engineer, Dughwede man, interviewed at Gwoza, 20th October 2007. He also said the name could also mean ‘land area.’
7 K M Buchanan & J C Pugh, Land and People in Nigeria: The Human Geography of Nigeria and it Environmental Background, London, University of London Press, 1955, 110
8 Gwimbe, Building Terraces in Agriculture, 1-2, regarding the Ngas side of the issue. During an interview with Musa Wahe, 55, Teacher, Guduf man; Habila Kulkwe, 62, Civil Servant/Farmer, Guduf man; and, Emmanuel Mbitsa, 55, Carpenter/Farmer, Guduf man, at New Guduf, in the foothills of Gwoza peninsula, on 21st October 2007, it was stated that the Zwagwada (or Guduf) people were said to have moved from Ngazargamu to the north, into the Cameruonian area, in the south, before relocating to their present area on the peninsula.
9 Abdullahi Smith, A Little New Light: Selected Historical Writings of Professor Abdullahi Smith, Zaria, The Abdullahi Smith Centre for Historical Research, 1987, 83
10 The southwest shores of the Mega Chad, the basis of the Chad Basin, rested more or less directly on the northern edge of the Jos Plateau, around Jos, thus slightly away from the borders of Yil Ngas lying to the southeast, albeit the latter is quite within its fringes. In defining the limits of the Chad Basin, Graham Connah gives this narration: “The Chad Basin, bounded by Air, Hoggar, Tibesti, Ennedi, Marra, Adamawa, Mandara highlands [Gwoza], and the Jos Plateau covers an area of about two million square kilometres.” See: Graham Connah, ‘Some Contributions of Archaeology to the Study of the History of Borno’ in: Bala Usman & Nur Alkali (eds. ), Studies in the History of Pre-Colonial Borno, Zaria, Northern Nigeria Publishing Company, 1983, 5
11 Keir Hansford, John Bendor-Samuel & Ron Stanford, Studies in Nigerian Languages, No. 5 An Index of Nigerian Languages, Accra, Summer Istitute of Linguistics, Ghana, 1976, 118.
12 This was later put in printed form in: C G Ames, Gazetteer of Plateau Province, London, Frank Cass, 1934, 22
13 Graham Connah, Three Thousand Years in Africa: Man and His Environment in the Lake Chad Region of Nigeria, Cambridge, Cambridge University Press, 1979, 18. Stanhope White maintains in: White, ‘Agricultural Economy of the Hill Pagans of Dikwa Emirate’ Farm and Forest, 130, that: “The peoples now dominant all affirm that they came from the south at some distant period…” without mentioning specifically which peoples. However local intra-Gwoza peninsula movements and migrations, even the colonial change of mandate, as it were, in 1919 from the German Cameruon more or less to the south, to the British Nigeria, more or less to the north, could partially have engendered a southern-pointing tradition, in the course of time. But Gwoza languages, are predominantly Laamang-Chadic, further supporting the northern-pointing traditions of origin.
14 Robert McC. Netting, Hill Farmers of Nigeria: Cultural Ecology of the Kofyar of the Jos Plateau, Seattle, University of Washington Press, 1970, 44
15 J M Kowal & A H Kassam, Agricultural Ecology of Savanna: A Study of West Africa, Oxford, Clarendon Press, 1978, 31
16 Buchanan & Pugh, Land and People in Nigeria,110; & J H Mackay, ‘Perspective in Land Planning: Part 2’, Farm and Forest, Vol. V, December 1944, 106
17 Kowal & Kassam, Agricultural Ecology of Savanna, 174
18 Buchanan & Pugh, Land and People in Nigeria, 109
19 Buchanan & Pugh, Land and People in Nigeria, 109
20 I have seen hillside farms of the Eggon, Mada, Mwaghavul, Pyem, Taroh and so on, where elaborate, if any, terraces are not in evidence.
21 Gwimbe, Roots of Rural Poverty in Yil Ngas of Plateau State, 38
22 White, ‘Agricultural Economy of the Hill Pagans of Dikwa Emirate’ Farm and Forest, 131: Though the figures suggested in the article are guessed, not based on research.
23 Buchanan & Pugh, Land and People in Nigeria, 111
24 Buchanan & Pugh, Land and People in Nigeria, 110
27 White, ‘Agricultural Economy of the Hill Pagans of Dikwa Emirate’ Farm and Forest, 131
28 During my visit to Gwoza, I heard of one Gehard Kosack, believed to be an archeologist or so, having in Gwoza , on research, from London, shortly before I got there. Time and effort did not permit me to try and contact him before completing this paper even though I was given his paper mail address.
29 White, ‘Agricultural Economy of the Hill Pagans of Dikwa Emirate’ Farm and Forest, 131
30 Buchanan & Pugh, Land and People in Nigeria, 110
31 Elizabeth Isichei (ed.), Studies in the History of Plateau State, Nigeria, London, Macmillan Press, 1982, 31-32
32 White, ‘Agricultural Economy of the Hill Pagans of Dikwa Emirate’ Farm and Forest, 131
33 Boyd Alexander, From the Niger to the Nile, Vol I, London, Edward Anold, 1907, 97
34 ‘Agriculture: Plain Land and Terrace Farming’ in: Rural Voices, A Rural Highlights Publication, January 2007, 8
35 McC.Netting, Hill Farmers of Nigeria, 61
36 Gwimbe, Roots of Rural Poverty in Yil Ngas of Plateau State, 75-77
37 White, ‘Agricultural Economy of the Hill Pagans of Dikwa Emirate’ Farm and Forest, 130: White also noticed and reports the surprising smoothness of the stones on the terraces and rocky hillside of Gwoza.
38 Buchanan & Pugh, Land and People in Nigeria, 110
39 Keith Eugene Hess, ‘The Most Advanced Agriculture in Nigeria’ in: Keith Eugene Hess (ed), MCC West Africa’s Environmental & Food Concerns Newsletter, No 9, April 1992, 10: Regarding the article’s surprising title, the author says in its opening sentence: “That was my immediate response to the question, ‘What do you think of Gwoza?’ It may be a bit of an exaggeration…” which simply paved the way, as shown in the citation, in the text of this endnote, for his personal opinion. In other words, the title was a product of a spur-of-the-moment response, not his deeply personal opinion, nor the product of a research.
40 Ishaya Haske, 23, photographer, my Gwoza escort, Mafa youth, interviewed at Guduf hills, 21st October 2007.
41 Gwimbe, Roots of Rural Poverty in Yil Ngas of Plateau State, 60-61 & Gwimbe, Building Terraces in Agriculture, 12
42 White, ‘Agricultural Economy of the Hill Pagans of Dikwa Emirate’ Farm and Forest, 131
43 Connah defines firki as “the clay plains to the west, south and east of the present lake”, meaning Lake Chad. See: Connah, ‘Some Contributions of Archaeology to the Study of the History of Borno’ in: Usman & Alkali, Studies in the History of Pre-Colonial Borno, 6. Firki agriculture, which involves the transplanting seedlings of the masokawa sorghum species between September and November, which florish only on reserved underground moisture, not on any rains which by then have normally ceased, to be harvested in either December or January following (an entirely ‘Ember’ months affair, more or less!), practised to an extent in the foothills and plains of Gwoza, though mostly practised in the Chadian firki fadamas, was, to me, an unimaginable, unanticipated, discovery. It seems to be a form of agricultural production with the least effort. Once the crops, with the aid of their shaft-like digging sticks, have already been transplanted in holes on the plain ground, the plots having hitherto been cleared by bush burning, the farmers would simply wait for harvest!
44 Ames, Gazetteer of Plateau Province, 9
45 Isichei (ed.), Studies in the History of Plateau State, Nigeria, 33-34: Isichei further maintains: “The kenti or Kyang obtained by the lowland Ngas from the Bauchi Kantana [who partially inhabit Kanam local Government Area of Plateau State, lying to the south-east of Yil Ngas] was five inches long and three and a half inches wide. The chen [or big hoe] made by the hill Ngas was much larger and spade-shaped.” Neighbors of the Ngas were and are sometimes surprised with the size of Ngas hoes.
46 Connah, ‘Some Contributions of Archaeology to the Study of the History of Borno’ in: Usman & Alkali, Studies in the History of Pre-Colonial Borno, 7
47 McC.Netting, Hill Farmers of Nigeria, 58
48 Peter Yilkur Daji, 50, Civil Servant: History Academic, Taroh man, phone-interviewed at Gindiri, 22nd October 2007
49 Hansford, Bendor-Samuel & Stanford, Studies in Nigerian Languages,153
50 Generally from informants at Gwoza, between 17th and 21st October 2007, including – among others, Rev Daniel Gula, 68, Retired Personage, Glavda man; Mrs Ruth Kathlene Gula, 67, Briton/Missionary, Married/Resident in Gwoza for more than 35 years; Nathan Habila, 41, Civil Servant: Local Government Works Engineer, Dughwede man; Ishaya Haske, 23, my Photographer/Gwoza Escort, Mafa youth; Musa Wahe, 55, Teacher, Guduf man; Habila Kulkwe, 62, Civil Servant/Farmer, Guduf man; Emmanuel Mbitsa, 55, Carpenter/Farmer, Guduf man; Umar Abba, 23, Applicant/Inn-keeper, Dughwede youth – the common position is that the present inhabitants are the descendants of the erstwhile terrace-builders. Even in 1992, the position was different from that of 1941, it was most possessive: “The people seem to think they have always lived on the mountain and that they always built terraces”! See: Hess, ‘The Most Advanced Agriculture in Nigeria’ in: Hess, MCC West Africa’s Environmental & Food Concerns Newsletter,10

Return to Table of Contents