The history of dagga in the Iron Age

I will be examining the history of Cannabis sativa (from here on, it will be referred to as dagga) in the Iron Age of Southern Africa. The main objective of this thesis is to clarify the approximate date that dagga entered into the continent, which group of people introduced it into the continent as well as what the main practices and uses were of dagga at that particular time.

I will mainly draw upon past and current literature regarding dagga. There is, however, a stone mortar and pestle that has been recovered from the Mapungubwe landscape and the aim is to test the stone to determine as to whether there are traces of cannabis on the stone as well as to determine an approximate date for the residue left on the stone.


Jan van Riebeeck

The aim of this honours project is to clarify the history of dagga in the South African archaeological record- with specific reference to the Mapungubwe landscape in Limpopo. Specifically, I will aim to determine whether dagga was introduced by the Bantu speakers who were migrating southwards or by the Indian Ocean Trade.

The diffusion of dagga into Africa- particularly southern Africa has long been contested and various researchers have offered different models as to the time of entry, and models of migration using a variety of sources and explanations.

Smoking in Africa has been of particular interest to archaeologists and historians because of the significant number of smoking pipes that have been found at various sites across the continent. Dagga smoking, among other practices is still prevalent in modern South African culture, thus it would be beneficial to explore the customs of the past to better understand the future

Literature Review
One of the first mentions of dagga in Africa was made by Friar João dos Santos in 1609 whereby he mentions that throughout Kaffraria in Ethiopia, the indigenous occupants grow a herb called bangue. He describes the plant as being similar to the ear of coriander in both the grain and ear but states that the leaf differs to that of coriander. He writes that the leaf is comparable to the clove gillyflower. Dos Santos explains how the natives dry the leaf and stalk and then grind them into a fine powder. From there, the powder is eaten and complemented with water. He states that the people who consume the bangue act in a drunk-like manner after eating it. Walton (1953) reiterates that dos Santos mentions the cultivation and use of dagga by the eastern Shona in the 16th century.

Perhaps one of the best ways to obtain dates for dagga use in Africa is by dating the pipes that were used to smoke dagga however, Jeffreys (1965) states that dating the pipes is rather perilous to date. This has now proven to be an overstatement as there are many dating techniques available. Jeffreys (1965) maintains that dagga smoking was a widespread practice in tropical Africa ± 1000 years before tobacco smoking was introduced to Africa. Walton (1953 as cited in Jeffreys 1965) believes that dagga was introduced into southern Africa by the first group of Bantu herders from the north and as a result, significant quantities of dagga pipe bowls have been found to be associated with early Bantu settlements in Zimbabwe, Gauteng, the Orange Free State, Basutoland, the Cape Province and Kwa-Zulu Natal. The pertinent question then becomes, at what time did the first Bantu speakers move into southern Africa. Breutz (1955 as cited in Jeffreys 1965) believes it was as far back as 1300 A.D. This date is clearly incorrect but it was the current thinking at that time.

Philips (1983) and van der Merwe (1975) offer that the first solid evidence for dagga smoking in Africa was unearthed by J.C. Dombrowski in 1971 in Lalibela in central Ethiopia. Dombrowski dated a number of artefacts from the surroundings stratigraphy layers where the smoking pipes were located and he concluded that dagga smoking was practiced in Ethiopia in the thirteenth or fourteenth century.

Two sites have been excavated in the Brandberg, Nambia which are associated with the Bergdama. These sites offer evidence for smoking. The radio carbon dates supplied for this site are 1590 and 170 A.D respectively. Based on ethnohistorical data, it has been suggested that the occupants of the afore mentioned sites were smoking dagga by the sixteenth century (du Toit 1980).

Four fragments of smoking pipes have been excavated at Sebanzi Hill in Zambia. The pipe fragments suggest that smoking in central Africa is of far greater antiquity that had been previously suggested. One of the pipe fragments has been tentatively dated to approximately 1200 A.D. whilst two of the other fragments have been dated to the seventeenth century based on artefacts located within the same stratigraphy- namely charcoal (Phillipson 1965).

Huffman et al. (2002) excavated a stone-lined mortar and a stone-edged circle at site 14C. Huufman (2002) believes that the stone-circle cannot be a fireplace as it is too high and too small. He believes it was probably used as a pot-stand. Through Huffman’s (2002) understanding of the Central Cattle Pattern, the location of the stone circle would lie in the men’s domain. Huffman (2002) suggests that both the location as well as the arrangement of features could indicate that the area was more than likely used to process cannabis into hashish. If this assumption is correct, then mortar was used to grind the cannabis into a pulp that was in turn placed into a calabash which then floated in a pot placed in the stone circle. The embers allowed for the necessary heat required to produce hashish which would, probably, be later smoked.

Du Toit (1980) states that dagga is social plant and is associated with human settlements; thus the spread of dagga was not due to wind, currents or animal movements. There are several current diffusion hypothesis models offered to explain to diffusion of dagga into Africa. Watt (1961 as cited in du Toit 1980) believes that dagga may have been introduced into Africa by early travellers circumnavigating the Cape from the east. The majority of all the historical documentation as well as the linguistic evidence suggests a date before the fifteenth or sixteenth century return of the European travellers.

James (1970 as cited in du Toit 1980) based his argument on single entity of terminological evidence- the relationship between the Hindi and Shangaan (Tsonga) languages. He maintains that the dagga was first introduced to the coast of Mozambique by the Portuguese militant traders who were returning from India. This model then offers a later date. James’s model is lacking as it does not acknowledge documents regarding the early uses of dagga.

Morley and Bensusan (1971 as cited in du Toit 1980) recognise that dagga is not indigenous to southern Africa. The authors believe that dagga was introduced into Africa by the Arab traders to the Mozambique coast from India. From Mozambique, the plant was carried towards the south by the migrating “Hottentots and Bantu” (du Toit 1980:15). This hypothesis is supported by Goodwin (1939 as cited in du Toit 1980). However, du Toit (1980) believes that the hypothesis is lacking in that it does not specify the details of the “Khoikhoi and Bantu” people. This hypothesis does, nonetheless, offer an earlier date for the introduction of cannabis.

Walton (1953 as cited in du Toit 1980) incorporates his own archaeological reports which refer to pipes found in early Bantu settlements as well as to Dos Santos‘ description of the cultivation of dagga by the eastern Shona in the sixteenth century. Walton believes that dagga was introduced into southern Africa by the first Bantu herders from the north. From there, the use of cannabis would have spread from the Bantu to the Khoikhoi and the San. Du Toit (1980) holds this hypothesis in high regard as it represents a large portion of the evidence he has managed to obtain.

Du Toit (1980) offers his own hypothesis that dagga spread from south Arabia through Ethiopia. It is well known that that the Amhara originated from Arabia and an array of commodities proceeded and followed this Semitic incursion- during this process plough agriculture, a zebu strain of cattle and various agriculture products spread to Ethiopia. Du Toit (1980) is unsure as to whether dagga was one of these products. Du Toit (1980) relies on the evidence supplied by Dombrowski (1971) and van der Merwe (1975) with regards to the pipes excavated in Lalibela in Ethiopia (as mentioned above). Du Toit (1980) concludes that these pipes imply that either dagga came into Ethiopia from southern Arabia or alternatively, it spread from the east African coast in a northerly direction from Bantu-speaking people to the Cushitic people. Du Toit (1980) highlights that one of the issues with this hypothesis is the fact that Lake Tana is in the central northern region of Ethiopia and that there is no significant evidence to say that there was a trade route between modern-day Kenya and Ethiopia.

Shaw (1938) states that dagga was introduced to the east coast of Africa by the Arabs and had spread to the Cape Peninsula by the time of van Riebeeck’s arrival.
A significant amount of evidence is based on linguistic evidence. There are two fundamental terms with regards to the history of dagga: bhanga which originates from Sanskrit which in turn resulted in the Hindi word bhang and kinnab which is an Arabic word. It is thought that kinnab may have lead to Linnaeus adopting the word for the suborder of Cannabis (du Toit 1980).

The early Arab traders introduced the term bang to Africa along with its on linguistic varieties. This term is found all over east and southern Africa. The origins of the term are listed as being hindi- bang, Arabic- banj and Persian- bandz (banj). In the east African region, south of Lake Victoria, dagga is known as bhangi (Kollmann 1899 as cited in du Toit 1980); perhaps due to early Swahili contacts. Further south, there are other variations of the term bhangi. The Thonga (Junod 1912 as cited in du Toit 1980) located in the Zambezi Valley refer to dagga as mbange whilst the Zimbabwean Shona community refer to it as mbanji. The Venda community, south of the Limpopo River refer to dagga as mbhanzhe whilst the Sotho speakers refer to dagga as mmoana or matakwane. A slight phonetic variation is noted among the Swazi-Zulu and Xhosa speakers who refer to dagga as ntsangu. The Lamba society in present-day Zambia refers to dagga as uluwangula (du Toit 1980). The above illustrates that dagga had been present in these societies long enough for these people to adapt the Hindi word, bang, to suit their own language.


African Language Dagga term Reference
East Africa bhangi Kollman 1899
Thonga mbange Junod 1912
Shona mbanji Hannan 1959
Venda mbhanzhe Warmelo 1989
Sotho mmoana/ matakwane Mabile, and Dieterlen 1950
Swazi-Zulu ntsangu Doke, Malcom, Sikakana and Vilakazi 1990
Lamba Uluwangula Du Toit 1980


The first use of the term dagga first appeared Jan van Riebeeck’s journal in 1658 although it was spelt ‘daccha’. However, it is almost certain that the plant is question is not actually dagga but rather Leonotis leonurus which is well known plant among the Khoikhoi. This plant is also referred to as Rooi dagga, Wilde dagga, Klipdagga which was smoked by the Khoikhoi instead of tobacco (du Toit 1980). There appears to be confusion in the early literature (especially within the primary sources) between cannabis and Leonotis leonurus. It appears that ‘daccha’ was the generally adopted term of all narcotics- both smoked and chewed.

Du Toit (1980) states that it is not possible to confuse the adult dagga plant with the adult plant of Leonotis leonurus because of the different flowers they each produce. Whilst the dagga plant produces a dull white flower, Leonotis leonurus produces a bright red flower. It is possible, however that due to the similar use of both plants as well its similar properties, some confusion may have occurred.

Nienaber (1963 as cited in du Toit 1980) has conducted the most complete linguistic study of the term dagga. Nienaber suggests two possible hypotheses by referring to the works of previous researchers. Hahn and Lichtenstein suggest that it possible that the dutch word for tobacco, tabak, which was habitually referred to as twak, was corrupted into twaga and later to toaga and finally resulted in the term dagga, Du Toit (1980) believes that this an implausible explanation.

Du Toit (1980) suggests that a more credible explanation is the KhoiKhoi word daXa-b or baXa-b which commonly refers to tobacco among other things, is the root noun from which the word dagga can be derived. When one specifically refers to the word dagga, one can note that the qualifier !am (green) being added to the root aforementioned, the result would be amaXa-b namely green tobacco or dagga. Branford (1978) has drawn similar conclusions- her study linguistic study revealed that dagga is probably derived from the Khoikhoi word daXa-b

However, many researchers such as Lichtenstein, Meinhof and Nienaber doubt that dagga is an original Khoikhoi term. Meinhof goes so far as to propose that dagga was originally a derivative of an Arabic word, duXan (tobacco). Du Toit (1980) states that no other language group in South Africa has ever used such a term or anything resembling it.

On the other hand, du Toit (1980) states that early European travellers in South Africa had trouble recording the terms they heard phonetically when interacting with the indigenous people. Thus, a variety of spellings is common when dealing with Khoikhoi words such as daccha (1658), dacha (1660), dackae (1708), tagga (1725), dacka (1775) and daga (1779). Since the early white settlers were introduced to dagga by the khoikhoi herders, it was only natural that dagga became the common term for cannabis (du Toit 1980).

I will be examining previous literature as well as site reports. From these I will be able to infer as to when dagga entered into southern Africa and whether or not they were smoking dagga or using it as snuff based on the distribution of smoking pipes.

If smoking pipes are found to be associated with early Iron Age sites, then it can be inferred that dagga was introduced by the Indian Ocean Trade. However, if dagga smoking pipes are found to be associated with Middle Iron Age sites, one can conclude that dagga was introduced by the migrating Bantu speakers.

The stoner mortar and pestle recovered from the Mapungubwe landscape will be sent to be tested to determine if there are any traces of dagga on the surface. If there are any surface traces, one can conclude that the stone mortar and pestle were being used to grind up the dagga to use as snuff or to chew it.

Reference List
Brandford, J. 1978. A Dictionary of South African English. Cape Town: Oxford University Press.
Doke, C.M.; Malcom, D.M.; Sikakana, J.M.A.; and Vilakzi, B.W. 1990. Zulu: insangu luluhungu English- Zulu, Zulu-English. Johannesburg: Wits University Press
Dos Santos, 1609. In Theal, G.1964. Records of South-Eastern Africa, volume vii. Cape Town: C Struik (Pty) ltd.
Du Toit, B. 1980. Cannabis in Africa. Netherlands: AA Balkema
Hannan, M. 1959. Standard Shona Dictionary (2nd ed.). Salisbury: The College Press
Huffman, T; Schoeman, M and Murimbika, M. 2002. Origins of the Mapungubwe Project- Progress Report 2002. Unpublished
Jeffreys, M. 1965. Smoking in Africa. The South African Archaeological Bulletin 20(77): 48
Mabille, A and H Dieterien. 1950. Southern Sotho English Dictionary. Basutoland: Morija Printing Work.
Philips, J. 1983. African Smoking Pipes. The Journal of African History 24(1): 303-319.
Phillipson, D.W. 1965. Early smoking pipes from Sebanzi Hill, Zambia. Arnoldia 40(1): 1-4
Shaw, M. 1938. Native Pipes and Smoking in South Africa. Annals of the South African Museum xxiv: 227- 302
Shaw, T. 1964. Smoking in Africa. The South African Archaeological Bulletin 19(75): 75-77
Van der Merwe, N. 1975. Cannabis smoking in 13th-14th Century Ethiopia: Chemical Evidence. In: Rubin, V (ed.) Cannabis and Culture: 77-80. Netherlands: Mounton & Co.
Van Warmelo, N.J. 1989. Venda Dictionary. Pretoria: J.L. van Schaik (Pty) ltd.
Walton, J. 1953. The dagga pipes of southern Africa. Researches of the National Museum 1(4): 85- 113


Name Amy Flatau
Student number


Supervisor Thomas Huffman
Title The history of dagga in the Iron Age


Leave a Reply

Your email address will not be published. Required fields are marked *