Chapter 4. Review of mixed farming systems

1. Review of the literature
2. Limitations of the literature
3. Scope and method
4. Output

In Chapter 2, seven alternative typological principles were reviewed: functional farming systems, economic specialisation movement patterns, livestock ratios, traction characteristics, crop-livestock interaction, and farming intensity. It was concluded that:

(1) no all purpose typology is likely to meet the requirements of ILCA's several research thrusts;

(2) many typological principles have the fatal flaw that supporting data on the required variables are either not available on a compatible basis or are insufficiently reliable for comparative purposes;

(3) that crop-livestock integration, and farming intensity, have the greatest theoretical and practical significance in relation to contemporary processes of change in semi-arid mixed farming systems;

(4) that crop-livestock integration, however, must be measured in terms of a range of variables, sometimes giving contradictory signals, and the assignation of an 'integration score', averaged across these variables, may not therefore have much practical usefulness;

(5) that farming intensity, which can be reduced (at some risk of oversimplification) to a single value for each system - the cultivated percentage - offers a taxonomic principle both readily measurable (from air photographs) and relevant to crucial issues of livestock management (grazing systems, feed resources, nutrient cycling in the farming system).

1. Review of the literature

An attempt has been made to review the available literature in French and English on mixed farming systems in the SAZ as defined for the purposes of this Study.

Constraints Two constraints were imposed on this review.

(1) Literature has been sought on all countries having substantial SAZ, rather than accepting the uneven distribution of available studies; this systematic objective has been only partially met since time did not permit exhaustive searches to be made. Angola and Mozambique, in particular, have not been adequately covered in the search.

(2) Studies have been included in the review only if they post-dated the droughts of the early 1970s, reflecting conditions during the last two decades when average rainfall has diminished by 30 per cent or more over a large part of the SAZ, compared with the means for 1931-60. This constraint implies that many of the 'classic' anthropological accounts of livestock-keeping societies give an unreliable guide to contemporary trends in management and in the environment (both natural and economic), an assumption that should not always pass without question, but which was applied throughout for the sake of consistency. The only exceptions were made for systems on which no recent characterisations could be found.

2. Limitations of the literature

The literature has several major limitations for the purpose of guiding research in the 1990s. Among these are the following:

(1) Uneven geographical distribution, as just mentioned. There are significant contrasts between (a) Portuguese and non-Portuguese speaking countries; (b) Anglophone and Francophone countries (in favour of the latter); (c) favourite and unpopular countries in each of these groups (for example, Chad and Tanzania have been relatively neglected; Senegal and Zimbabwe on the other hand enjoy numerous recent, rigorous studies); (d) favoured and neglected regions or societies within individual countries.

(2) Variability in research objectives. A variety of research questions has been asked, reflecting the variety of professional disciplines involved. Consequently, the system characterisations have limited compatibility, and many questions were ignored if not seen to be relevant to the authors' stated objectives. This point is significant from ILCA's point of view since research objectives have been more rigorously defined since it came onto the scene, yet not much literature (outside ILCA's own substantial output) reflects these redefined objectives.

(3) Mixed farming systems were neglected until recently, in favour of specialist livestock or crop producing systems. Where crop producing systems had a significant livestock component (most often small ruminants), it tended to be treated as marginal to the cropping enterprises, like farm forestry, which also plays a significant role in some farming systems. Explicit attention to, and attempts to quantify, the linkages between crops, livestock and trees has not been characteristic of the bulk of the literature on African farming systems.

3. Scope and method

The review was exploratory in nature and designed to discover whether a basis exists within the literature on SAZ mixed farming systems for a typology. Two options were available: (a) to concentrate on a small number of systems (say 10) and review the literature on those systems in depth; or (b) to search widely for compatible characterisations if at a more superficial level, of (say) 50 systems. The second option was preferred because of the known diversity of the SAZ and its farming systems, and the lack of a principle on which the selection of a small number (option a) could be based.

The review comprised the following stages:

(1) Bibliographical search. A total of 500 references to potential case studies were listed from available sources in English and French and where possible scanned or abstracted. These include published items (books, journals), consultancy reports, theses, government documents and those of international organizations. It is probable that the items listed represent a fraction, perhaps a half or two thirds, of the materials in existence in a diversity of locations. Only about 30 percent of the listed items have been seen, however, and since their titles are rarely a reliable guide to the presence or absence of usable system characterisations it is not possible to estimate the value of the unreviewed literature for the purpose of the present study.

(2) Case study review. 65 items were reviewed, representing 30 per cent of the listed references. Of these 11 were subsequently rejected either because the systems described fall outside the SAZ as redefined (see Chapter 3) or because they contained insufficient information. A further 12 were merged with other studies of the same systems or areas, and one was split. The resulting 43 case studies were reviewed under 32 standard typological variables. The list of case studies is given below.

List of case studies reviewed

Number	Country	Ethnic Group (major) or area
01	Mali	Bambara
02	Nigeria	Hausa (Kano Close-Settled Zone)
03	Nigeria	Manga
04	Botswana	Tswana
05	Niger	Hausa (Maradi)
06	Cameroun	Mafa (Mandara)
07	Sudan	Gezira (Arid Zone: see Appendix 5)
08	Sudan	Lahawin
09	Senegal	Serer
10	Mauritania	Soninke
12	Burkina Faso	Mossi
13	Burkina Faso	Tuareg
19	Somalia	Somali (irrigation)
21	Somalia	Somali (agro-pastoral)
22	Kenya	Akamba (Machakos)
23	Tanzania	Hehe
28	Ethiopia	Beni Amer
29	Ethiopia	Dassenich (Geleb)
30	Ethiopia	Oromo
33	Ethiopia	Hamar
37	Ethiopia	Arsi
39	Ethiopia	Tigreans
41	Tanzania	Barbaig
43	Ethiopia	Somali
44	Mozambique	(south)
46	Zimbabwe	Ndabele
47	Zimbabwe	Shona
48a	Zimbabwe	Ndabele
48b	Zimbabwe	Shona
49	Kenya	Il Chamus
50	Kenya	Akamba (S. Kitui)
51	Kenya	Akamba (S. Machakos)
53	Kenya	Maasai (Kajiado)
54	Lesotho	Basotho
55	Angola	Khumbi
57	Botswana	Ngamiland
59	Botswana	Bakalanga
60	Sudan	Baggara (Hawazma)
61	Sudan	Baggara
62	Sudan	Nuba
63	Zimbabwe	Shona
64	Senegal	Wolof
65	Ethiopia	Eritreans, Tigreans

The locations of these case studies are shown on Figure 6.

(3) Reformatting. During the course of the review it became apparent that some of the variables could be discarded without loss, and others condensed, for the typological objective in view. Accordingly the 43 case studies were reformatted on 32 variables, which include scores on 8 variables of crop-livestock integration (nos. 17-24), and a score for farming intensity (no. 29). The list of variables is as follows:

Number	Head	Code
1		Case study number(s)
2		Source(s)
	DESCRIPTORS
3		Country, locality
4		Rainfall, environmental unit, and strata
5		Ethnic group(s)
6		Critical ecological indicators
7		Human population, density, growth
8		Livestock population, density, growth
	RESOURCE ACCESS
9		Livestock/holding - types, numbers
10		Livestock ownership determinants
11		Access rights - grazing
12		Access rights - farmland
	ECONOMIC INTEGRATION
13		Contribution to subsistence
14		Contribution to income
15		Investment value
16		Exchange contracts.
	SYSTEM INTEGRATION
17		Residues
18		Fodder trees
19		Fodder production
20		Manure
21		Traction
22		Transport
23		Cattle movements
24		SR movements
	RECENT TRENDS, ECONOMIC
25		Settlements
26		Land supply
27		Specialisation, diversification
28		Market impact, terms of trade
	RECENT TRENDS, ENVIRONMENTAL
29		Intensity rating
30		New systems of resource use
31		Degradation, sustainability
32		Effects of drought

4. Output

The output, in the form of standardised summaries of the case studies, is presented in Appendix 1.

The incompleteness of many of the entries will be apparent. This reflects the inadequacies of the sources (for this purpose). Many of these gaps could be filled from further searches in the literature. The present operation was severely constrained by the time available, and its purpose is illustrative rather than definitive. Enough has been done to show the potential and the limitations of this type of approach to classifying the literature.

(1) It provides a systematic method of abstracting compatible data at low cost from the existing literature and maximising its value for the purpose of targeting research, identifying recommendation domains, etc.

(2) It provides a method of identifying the gaps both in geographical coverage and in knowledge.

(3) It offers a basis for an ongoing inventory of mixed farming systems, using ILCA's in-house resources and a sharpened or modified variable 'menu'. Such an inventory may have value to other agencies interested in livestock research and in dryland management.

(4) On the other hand, such an approach can be no better than the literature on which it is based.

(5) It cannot provide an input to specific research programmes or substitute for specialised literature searches. Its purpose is restricted to the typological or taxonomic objective.