\ econa internaiionai seminar
on Change in Agriculture
91 Reading England 9-19 September 1974
THE CHOICE OF METHODS FOR IIPLEMENTATION:
By Guy Hunter
Overseas Development Institute, London
1. One basis upon which much of the programme of this Seminar, and
the choice of background Papers, has rested is the work of the Reading
University/Overseas Development Institute joint programme of research,
which has been running for the last five years. The research was
based upon a belief that, however good agricultural policies might be,
there is a widespread failure in implementation; and that a major part
of this failure (certainly, not all) could be ascribed to a failure to
learn the lessons of experience in the choice of organisational methods
and of institutional forms. The programme was run in what was, in
1969, an unusual method. The field research financed by the programme
(which was in turn financed by O.D.M., the Ford and Rockefeller Founda-
tions, F.F.H.C., Barclays International and Shell International) was
executed, save in one instance, by Indian and African scholars, in con-
sultation with the Reading/O.D.I. staff; and- its initiation was the
outcome of consultation with the Ministries of Agriculture in India,
Kenya and Nigeria. The work was supplemented by 'library work' in
London, the overseas experience of the staff and a few specially
2. The purpose of this present Paper is to lay before the Seminar
the main hypothesis underlying this work. It does not cover the whole
range of subjects to be covered by the Seminar. This hypothesis under-
lies both the research, the choice of documentation, the subjects of
plenary addresses though their content is not dictated! and the
arrangement of subject matter.
3. The hypothesis .can be put in the following form: "Agricultural
development takes place among local farming communities at various points
in a continuous transition from fully 'traditional' to more 'modern'
technical and social organisation; in various different ecologies; with
various post-harvest treatment of crops or animals produced; and under
the guidance of Governments with different resources of personnel, ad-
ministrative capacity, and budgetary resources. In considering the
choice of methods (administration, organisation and institutions) for
implementation of agricultural policies, consideration of four main
factors will lead to choices which are more likely to succeed -
1) The attitudes, capacities and needs of the local
farming community at the time.
2) Technical factors, especially the type of crop
or animal husbandry mainly concerned.
Choice of Methods for Implementation
3) The nature of the processing and marketing channel.
4) The administrative resources and capacity of the main
agency of change (usually, the Government or a
4. These criteria may look deceptively, and indeed unrealistic-
ally short and simple. In such an immensely complex subject they cer-
tainly cannot be complete and decisive. It is necessary, therefore, to
spell out some of the major issues which underlie each 'criterion' -
perhaps 'guideline' would be more modest and this is done below. We
do not claim that the use of these criteria will provide to overseas
Governments or donors an infallible and precise guide to the choice of
administrative methods and institutions in all cases; the world, and
chance, are too complex for that. But we do claim that if the criteria
are carefully considered and sensibly used, the repetition of grave mis- '
takes will be substantially reduced, and the chances they are still
chances of success will be substantially, and even critically, increased.
Criterion 1: The Attitudes, Capacities, and Needs of Farmers
5. This criterion is by far the most difficult and complex to apply.
In commonsense terms, backward 'tribal' farmers in India or Africa will
differ in attitudes to innovation, in their capacities to manage change,
and the need of assistance which they have, from a group of sophisti-
cated Punjabi wheat farmers, or Kilimanjaro coffee farmers, or success-
ful West African cocoa farmers, who have long since learned to adopt
scientific methods, geared the farm to cash.earnings, and learned to
adapt to both market prices and new technological advances. But between
these extremes lie the majority of situations, where some change in
attitudes and capacities has taken place, but, in varying degrees, not a
complete change and one which is not yet self-assured and secure.
6. Attitudes may be religious, social and customary, and economic.
They tend to be highly specific in place and time, particularly in the "
earliest stages of innovations, when local sub-cultures retain maximum
vigour. There are, however, a few general attitudes which are widespread
at this stage aversion to taking risks with the main subsistence crop;
fear and suspicion of outsiders, especially officials; dependence of small
farmers on some forms of power or authority within the community public
opinion of the collectivity itself, chief, landlord, etc. Beyond these
generalised fears there may be far more specific rules and taboos about
particular crops (especially the staple food) as to planting dates, etc.,
and communal arrangements for such matters as access to fields, mutual
help, grazing rights. In general, the process of modernisation, inclu-
ding wider contact outside the village, involves a gradually increased
dominance of economic motives, at the expense of attitudes which obstruct
economic success, and a corresponding diminution of some (but not all) of
the attitudes and behaviour patterns of the traditional idiosyncratic cul-
7. Capacities. The technical skill of the farmer himself is not
usually a critical issue he is in many ways skilled already, and can pick
up line-sowing, for example, in a season or two. Illiteracy is, of course,
limiting at later stages, when chemical and engineering techniques, and
*' T r~. -. . -:i .. c ).1.- r., ,.,, r.^f- \, ,
Paper le: 3
can be critical in innovation; but this is more relevant.to the
content of the 'package' and less to organisation and institutions.
Socially determined capacities are n:ore important. The farmer, for
many reasons (tenure, public opinion, etc.) may not.be able to use his
land as he might wish, whatever his personal attitude. Economic
capacity of the farm family (apart from labour) may be restrictive. A
short-straw variety may reduce necessary thatching or animal foodstuff,
may deprive the family economy of vital milk or ghee or transport or
fertilizer; high cash input requirements (even on credit) may be beyond
him, if the cash-flow in the rural economy is minimal and indebtedness
gravely feared or socially disapproved. Finally, managerial capacity
in a social context (e.g. active participation in running a Cooperative)
may be very low, not merely from lack of sophistication in handling money
and accounts but because of social dependency which excludes challenge to
8. Needs, naturally, overlap lack ofcapacities. Access to inputs,
to markets, to water supply, to fencing material, to information are
among the most common needs, which must usually be supplied organisa-
tionally from outside before adoption of new methods can even start.
Extension services can be greatly frustrated by the lack of investment
and organisational services outside their control but vital to the pro-
gramme which they are trying to achieve.
9. Implications for organisation flowing from this and other criteria
are dealt'with below.
Criterion 2: Technical and Environmental
10. Ecology, population density, type of human settlement, type of
crop grown, seasonality, will all, in varying degrees, affect the
organisation of implementation. Certain crops and organisational
requirements e.g. a steady flow of uniform-sized and equally ripe
tomatoes for canning --imply performance which may be extremely diffi-
cult for small farmers. Irrigation involves organisation and disci-
plines which may be new; extensive pastoralism involves considerable
modifications of extension pattern and marketing organisation; tree
crops with a long period before bearing involve investment in labour
and possibly credit or even subsistence payments. Acute shortages of
resources land, grazing, water, supplies will involve government
action to ration or control. Highly dispersed settlement patterns
make grouping (Cooperatives etc.) hard to organise, in contrast to
dense and compact communities. lost of these points are painfully
obvious; but they are not always taken into account, particularly
where the organisation of programmes, extension staffing and credit
systems are highly centralised and governed by uniform rules: to
contact 400 farm families in a big village might be possible for one
extension officer; the same number in a pastoral area might be quite
Criterion 3: Nature of Processing and Marketing Channel
11.' This could be regarded as an extension of Criterion 2
(Technical). It includes the cases where a crop (Tea, Tobacco, Sugar,
Rubber, Palm-oil are examples) requires major processing for a market
well outside the village (domestic or international 'export'). Such
Choice of Methods for Implementation
cases lend themselves to 'integrated management' by a Company, very large
Cooperative, or parastatal Board, servicing out-growers, and often pro-
viding research, Extension, credit, collection of the crop, processing,
grading, marketing and payment. There are many well-known examples.
This system supplies from outside the managerial capacity which farmers
in early stages may lack, whether as individuals or as Cooperatives, and
can sometimes be applied at iery early stages of modernisation.
12. In contrast, staple cereal foods, which may be used: a) for
family consumption; b) for paying share-crop rents; c) for brewing beer;
d) for repaying obligations; 'e) for seed and insurance-storing, and
f) for minor sales, present a far more difficult problem. For. example,
crop-season credit for fertilizer, which will be applied to the whole
crop, may be very hard to recover because: a) the credit agency does
not control the disposal of the crop; b) the credit debt looks very
high in relation to the few bags which may ultimately be sold for cash;
and c) because small merchants usually have an advantage over official
organizations in handling this type of situation.
Criterion 4: Administrative Capacity of Government
13. This poses difficult issues which have rarely been raised in
this form. Because government disposes of a number of paid staff
S(Extension, Cooperative, Community Development, Credit service, etc.,
etc.), it is easy to.draw up on paper increasingly elaborate duties and
increasingly sophisticated systems of coordination (for 'integrated
rural development', for example), as though the staff were a totally
flexible instrument of infinite capacity. 1loreover, as the years go
by, more duties and more committees and reporting requirements are added,
without reducing the original load. A second, common, phenomenon is a
tendency to create a new Agency for each new need, so that Crop Boards,
Land Boards, Credit Corporations, Natural Resource Boards, Irrigation
Authorities, Ranching Corporations, Settlement Boards, with their in-
evitable accompaniment of coordinating committees, are set up in pro-
fusion, and almost without regard either to the skilled manpower available
(especially at field level) or to the conflicts of function and authority
which, though excluded on organisational charts, invariably arise in
14. It would seem clear that the complexity of organisation must be
related to the skill-resources of government. Where these are small,
extremely simple organisational forms, with.a high degree of delegation
of discretion to act locally, will be necessary. This cuts across the
manifest tendency to increase complexity and centralisation through
planning controls, theoretical perfectionism, attempts to do too
many things at once (very evident in 'integrated' schemes), and a blind
eye to the realities of Departmental jealousies, bureaucratic traditions
and the motivation, conditions of service and career opportunities for
the minor staff in the field.
Limitation of the Hypothesis
15. The focus of the Reading/ODI work has been on the choice of
organisational and institutional forms within a given policy. Thus
the field is limited in several ways. First, it is not.directly con-
cerned with the formulation of the content of policy, except, somewhat
Paper le: 5
indirectly, in so far as a policy may be organisationally impossible to
implement. Policies may be chosen primarily on political grounds; and
political beliefs may even rule out certain organisational methods -
e.g. private enterprise, though it will still have to find a mechanism
of implementation, largely through public institutions: a bureaucracy,
Party Cadres, Cooperatives, Communes, etc. In this Seminar, the
opening sessions take account of some of these political choices.
Secondly, technical agricultural policy was outside the field cf re-
search. Clearly, if a policy of growing cotton is applied to land
which is unsuitable there are many lesssimple but still catastrophic
mistakes no amount of good organisation can prevent failure. To a
large extent technical choices are.very highly location-specific, and
therefore unsuited for handling in generalisations of the type.which we
have considered. Thirdly, the research has concentrated on develop-
ment policy for the mass of small farmers, largely neglecting planta-
tion agriculture and (except by implication) the management of very
large collective or State farms.
16. By this time we feel able to go beyond the mere statement that
the four Criteria must be 'taken into account', and to suggest, from
the basis of experience and research, some at least of the detailed
implications for organisational choices which flow from the use of these
Criteria. These implications are arranged under organisational head-
ings, since they may result from applying more than one of the main
criteria. I have largely avoided using a 'stage-theory' presentation,
because of the well-known difficulties of such an approach& But
because we are dealing with a transition, through time, from one pattern
of agricultural and economic activity to a different pattern, and
because organisational choices have to take into account the point on
the line of transition which. a particular farming community has reached
at a given time, the concept of sequence and timing through the
'stages' of transition, although they are blurred at the edges, will be
always in evidence.
17. Because of the insecurity, suspicion and fear of change which
is strongest in the very earliest stages of development, it would
appear that a classical 'Community Development' approach is best suited.
to communities at this point. Ideally, C.D. staff are trained to get
to know a farming community, to listen, to help the community to meet
expressed needs, and thereby to gain the trust of the farmers. They
may go beyond this, in suggesting new possibilities (not locally known,
and therefore not felt2as needs), but they will. not act on these
without local consent. This style contrasts, unfortunately, with
the most common style of extension staff, who come with a package of
supposedly superior practices, usually centrally devised, which they
try to persuade the community to adopt.
18. If this initial contact is successful, a time will come when
the farmers, or a substantial number of them, want to go a step
further, particularly in increasing incomes rather than in improving
social facilities, with which C.D. is often most concerned.
Choice of Methods for Implementation
At this point fairly simple but well-founded technical agricultural
advice, through extension of tested agricultural techniques, becomes
appropriate. There are occasions when a complete and advanced agri-
cultural package can be introduced 'at a blow'; "but since this may
involve multiple changes, perhaps commercial as well as technical
(purchased inputs, credit, new types of organisation, as well as a dif-
ferent agronomy and crop-protection system), it is more often desirable
to advance by stages: line-sowing before fertilizer, savings before
credit loans, improved cultural practices before revolutionary changes
in varieties and methods. Close contact with farmers by relatively
simply trained staff will here be essential.
19. At a much more advanced stage the role of extension and the type
of staff.may have to change. The period of motivating farmers to accept
change will have passed; what they now need is more and more specialised
advice, both agronomic fertilizerr mix, disease control, implements) and
economic (costs, prices, farm management). They will also require com- .
mercially significant quantities of inputs efficiently delivered and of
reliable quality. Extension staff will require better technical train-
ing, better supported by specialist advice. The University may well
enter the extension field at this point.. Further, Government, quite
apart from the extension service, will face new duties, in the efficient
organisation of supply, repair, marketing, seed-production, agricultural
chemicals and machines. Further, because demand is rising, Government
may soon be able to pass over the executive responsibility (though not
the supervision) to the private sector, since the farmer has, at. last,
become a profitable customer for inputs and a producer of commercially.
worthwhile outputs, increased-both in volume and quality.
20. There is considerable evidence that, despite appearances, even
poor farmers can find sources of small amounts of cash (US $ 50) when
they really want to (e.g. for school fdes, to meet social imperatives).,
Secondly, that, despite appearances, savings groups of various kinds are
quite widespread in many traditional economies, and can also be stimulated.
Thirdly, that borrowers are prepared to accept quite high interest rates
on small loans for short periods, e.g. 10 per cent for 6 months, equi-
valent to a 20 per cent per annum rate (but-only $2.50 on $50 for six
months). Fourthly, that official crop-season subsidized credit schemes,
for farmers in a fairly early stage of development, are extremely costly
and not often efficient agriculturally. They are costly either because
of low repayment levels or (more frequently, nowadays) because of the
high staff costs of loan recovery, except in certain cases where credit
is given and recovered by an organisation having monopoly control of the
crop. They are agriculturally ineffective because the purchased inputs
are spread too thinly, or used for other than the intended purpose, or
partially used for consumption or social needs.
21. All these findings point to a far more cautious and more selective
use of official crop-season credit schemes. Some suggestions would be:-
a) To exhaust other methods of assisting farmers before
purchased inputs for credit started.
b) To stimulate savings before loans (the Comilla principle).
Paper le: 7
c) To stress cash purchases by emphasising the cost of credit,
which should be reflected in realistic interest rates.
d) Experimental use of free fertilizer for demonstration over
.1 year, followed by cash sales.
e) Encouragement of small, mutually guaranteeing, credit groups.
f) Use of Cooperative credit only after the Cooperative is
firmly established with adequate staffing and management,
since credit is both the least profitable and the most
difficult Cooperative function.
g) Extreme caution in credit provision where the destination
of the crop is not controlled by the credit agency.
22, All these suggestions reflect the fact that the wise handling
of production credit by farmers is a skill which comes late in his
growth towards modern farming, requiring training in its disciplines;
that institutional credit is not necessarily his first need, but may
become more important when he is already successful and wishing to
expand; and that, on the record, credit administration through
official channels is always difficult, and usually both inefficient and
Grouping of Farmers Cooperatives and Other Groups
23. Agricultural administration has to find some intermediary
between official services and the vast multitude of small farmers,
as a point for distribution of physical or credit inputs, as a channel
for distribution of information, and as a focus for shared facilities
(e.g. storage). Formal Cooperatives have been widely used for this
purpose, sometimes for political reasons, soiietimes simply as a con-
venient administrative tool. By criterion 1 (Attitudes, Capacities
and Needs) the circumstances in which a formal Cooperative is likely
to succeed in the various tasks set it are fairly few and specialised.
24. The Cooperative is a social organisation which cuts across the
most common forms of social grouping in most traditional agricultural
communities in the developing countries for example, kinship systems,
age-grade systems, landlord-tenant relations, patron-client relations,
employer-employee relations, clan systems, tribal societies (West
Africa), caste and status systems, and even the mutual assistance'
schemes (house-building, weeding, harvesting) common in many early
societies, which are built on wholly different lines.
25. Nevertheless, if the Cooperative is accepted as an alien
innovation useful for agricultural progress, we must estimate its
chances of success in performing two very different and distinguish-
able functions -
a) As a democratic, egalitarian system it is unlikely to
succeed in the early stages of development, when
attitudes of dependency are very high. Only after
a period of economic success which has included a
substantial proportion of previously dependent members
of the community are they likely to modify or sup-
Choice of Methods for Implementation
plant the dominance of traditional magnates in the
management of Cooperative affairs.
b) As an economic organisation cooperatives demand considerable
managerial skill, and a value system which puts a neutral
role efficiency above the obligations to political,
kinship or patron interests. Here again, a fairly late
stage of development is implied.
c) Purely as a coherence system (i.e. one relying heavily on
loyalty to the group), Cooperatives are likely to succeed
when the group is small, its members know each other
and have interests consciously shared. This would argue
that Cooperatives should initially be small (50 100
members). This conflicts with commercial efficiency and
capital-accumulation arguments, which point to large
societies with substantial turnover and capital, able to
provide worthwhile services. This difficulty may be
resolved by a small start, and a very gradual expansion.
d) The implications are -
i) that formal Cooperatives are not a tool of first
choice in the earliest stages:
ii) that coherence will be more likely if: a) the group
is initially small, and b) it is built round a
clearly needed physical facility (stores, pump,
well, motorboat, dairy, etc.) used equally by
all members. Pure credit cooperatives have the
least impetus to coherence each man wants his
26. There is an increasing emphasis on various forms of elective or
semi-elective popular representation, as an active element in agri-
cultural development, often including executive, or at least decision-.
making, responsibility. While these units at village-level (Gram pan-
chayat, Village Development Committee, etc.) have a fairly obvious func-
tion (to express local wishes and to contribute local knowledge), the
two or more higher tiers, which exist in many countries and in some
Projects, appear to have representative functions (they are no longer
face-to-face with village people), and in some cases executive functions
where staff are more or less directly under their control.
27. It is doubtful how far the Reading/ODI work can rightly include
this subject, since the purpose of establishing these Committee systems
is primarily for political education of the citizens and perhaps for
strengthening a dominant political Party by diffusing Party activity
widely through the provinces. But-in so far as development as such
is in issue, two points emerge. First, direct mobilisation of effort
is likely to be successful primarily at village level. The higher
levels give orders or pep-talks to villagers; but they may also have
an effect (where this is necessary) in prodding the executive bureau-
cracy into greater energy where that bureaucracy is controlled by the
central government. There is little evidence that the higher levels,
Paper le: 9
where they themselves control development staff, achieve high levels
of efficiency or impartiality. Secondly, in so far as development
depends on technical expertise and technical decision-making, there is
reason to fear failure. The record is often of mainly politically-
based decisions, and of ambitious projects which lack staff and ex-
pertise for implementation. In countries where trained personnel are
scarce, a tough and competent administration, prodded and checked by
local councils but not controlled by them, may be both more economical
in staff and more technically sound in programming. The information
available from mainland China would appear to contradict this statement,
since the Communes appear to be effective. But it is as yet hard to
distinguish how much this efficiency is owed to local election and how
much to a bold decentralisation of administration through the Party
cadres and nominees, combined with revolutionary enthusiasm and revo-
lutionary discipline too.
The Commercial Function
28. The variations in political policy and in the facts of national
history make this subject the least amenable to wide generalisation.
Some countries have indigenous traders and entrepreneurs; in some,
immigrants (Asians in East Africa: overseas Chinese in much of S.E.
Asia) have, unless politically excluded or restrained, pre-empted
much of the commercial sector. In heavily planned economies, and
anti-capitalist economies, or where there is no effective indigenous
trading network, this is the stamping ground for parastatal Boards and
Corporations, or state-supported Cooperatives.
29. Where these large organizations deal with a major crop of high
value, not mainly locally consumed in unprocessed or lightly processed
.form, with a fair proportion grown by sizeable and efficient growers,
they can succeed fairly well witness some of the Kenya Crop Boards,
originally aimed mainly at European growers. But faced by a mass of
small growers, bad access by road, uneven quality, and local markets,
they seldom can compete with small traders; and, if they have a mono-
poly, smuggling and black markets will appear, because (in contradiction
of the exploitative trader theory) traders and smugglers give the
farmers either better prices or quicker and more local service. There
is here a penalty exacted by ideological preferences which falls most
sharply on the small men whom ideology is designed to protect: it
may be a penalty outweighed by other political and social benefits. It
is also necessary to weigh the opportunities for patronage and cor-
ruption which Boards give, their re-emphasis on State and centralised
power, and the economic prizes they offer to the elites and the party
which can capture and monopolise control of Government.
30. The main implication has already been mentioned under Criterion
4 the necessity to match administrative patterns to available mana-
gerial skill and budgetary resources. What the Reading/O.D.I. pro-
gramme has not studied is the possibility of improving efficiency by
better management practices, ably set out in thepaper by Belshaw (5a),
and certainly of major importance. A glance at the administrative
superstructure of very small States will at once raise questions of
proportion. Can States with a population no bigger than that of one
or two Indian Districts really require so many Ministries, Departments
Choice of Methods for Implementation
DIFFICULTIES OF INTERPRETATION
31. Every person, every village is, in some degree, unique. Clearly
no administrative system can treat everyone differently. This is a
difficulty more real in theory than in (possible) practice. General
patterns of farm systems exist over areas at least as big as one extension
officer's area, and frequently to sizeable administrative areas. The
difficulty of adjustment lies in the administration, not in the facts.
Again, within a single village there may be a few sophisticated and wealthy
farmers, some halfway in transition, some still highly traditional -
who is to be served? The answer is again fairly clear the smaller,
less favoured, more 'traditional'. For if Government programmes are
firmly aimed to be feasible and profitable to them, the better-heeled
citizens will look after themselves. The trouble with much of the Green
Revolution has been that a fairly capital-intensive and complex package
has been offered which is, in effect, out of reach of the weaker members.
32. Again, how is the judgement the application of criteria to
be made? Can anyone weigh up all the factors, or decide at which point
in a complex transition one particular farming community stands, at a
point in time? Again, real life is simpler than theory. Really local
people know most of the real local facts, which seem so complex to out-
side observers, and which Governments at the centre rarely bother about.
What is, indeed, more difficult is to estimate the exact point at which
a local custom will bow to an economic incentive, and the exact moment
when new local leadership will emerge. These.questions can indeed only
be answered by (intelligent) trial and error.
335. If the need for closer adaptation of policies and programmes to
local situations is accepted, the clear implication is that these situa-
tions must be 'known', and the knowledge acted upon. They are, of
course, known by the people who live in them. 'Known' must mean known
to government, administrators, decision-makers; and also, perhaps,
'analysed, quantified and recorded', since this is the language which
officials and planners understand. This leads to the major, central
implication: that local programming decisions must be made very near to
the field; because it is only there that there is any real chance of
effective local knowledge.
34. Four processes are necessary to push decision-making and dis-
cretion downwards from the centre; to establish an acceptable point to
which it is pushed; to establish an effective contact with farmers and
an'.upward flow of information from them; and to retrain field staff to
listen first and advise afterwards. Everyone knows how difficult this
is, but primarily because the first step delegation of authority is
never taken. It is not t;:ken because: a) politicians, planners and
administrators at the centre insist on knowing best; b) simultaneously,
knowing that they don't really know, they hang on to slowly changing
generalised orthodoxies: 'Credit is the first step', 'Cooperatives must
be created', 'Elected committees must be set up everywhere', 'Traders
are exploiters', 'Extension staff must deliver packages and achieve
targets', 'Integrated Rural Development'.
35-. Ministers, planners, donors and universities all share the
blame for these orthodoxies. There are, indeed, some hopeful signs
of change. 'District Planning', 'Farmer Service Centres' at local
levels; some variations on the Cooperative model. But they are still
tentative; and there is still a great deal of detailed work to do in
establishing the minimum essential central control; the maximum feasible
local discretion; the point and the quality of expert technical input;
and the variety of forms of farmer organisation and contact. It is not
only work which is needed, but a change of heart at the centre, both as
to exercise of authority and as to the nature of the agricultural plan-
1. The paper by Phillips and Collinson (4c) and the chapter by
Waheeduddin Khan in Serving the Small Farmer refer.
2. See, e.g., the Animation approach in Niger described in the paper
by Gentil (3o).
3. This transition is admirably described in Kahlon's chapter in
Serving the Small Farmer.
4. Papers by Youngjohns (3i), Hyden (3k) and Texier (3p) refer
5. The paper by Haldipur.(Sb) and the chapter by Sinha and Jain
in Serving the Small Farmer refer.
6. Here Trapman's report on Kenya is highly relevant.
7. Roughly 1- million inhabitants to one District.
SOCIAL SCIENCE AND DEVELOPMENT
ANALYSIS AND ACTION
by Guy Hunter
April 15, 1975
1. One major danger in the use of social sciences in development is
that policy and action should follow too closely the findings of partial
2. This danger is visible at all levels from general ideologies to
micro decisions. At the macro-level, Marxism is as good an example as any.
Over the last century Marxist theory has had a compelling appeal, particularly
to intellectuals. Its sweeping analysis, based on a high degree of abstrac-
tion from selected processes of history and applied to a huge range of human
activity politics, economics, culture, religion has a satisfactory ele-
gance and comprehensiveness. But when the believers try to guide action from
this analysis, they have to explain away formidable anomalies, which arise
from factors assumed away in the analysis. The poor, instead of growing
poorer, grow demonstrably richer; well meaning reformers, of the type of, say,
Robert Owen, have to be dubbed as Menshevik enemies of the working man; colo-
nial District Commissioners, trying to persuade Africans to send their children
to school, or to plant corn in rows, have to be described as capitalist and im-
perialist exploiters. Worst of all, the analysis precluded any objective assess-
ment of how, after the revolution, the new governors would avoid the temptations
. of power and greed, or solve the social tensions aroused by a philosophy of
violent group conflict needed to heat up the revolution; nor did it consider
the social and economic results of attempting to run large nations through the
clumsy bureaucracy of the State.
3. At an intermediate level, economic Planning is subject to the same
criticism. The high degree of abstraction (from very suspect data) which is
involved in a national economic plan constantly neglects vast areas of human
motivation, ignores deeply held beliefs and fears, and, equally, forgets the
opportunities for evading or exploiting the mass of government regulations
through which its authors hope to ensure that their projections are fulfilled.
It may honestly confess the partial nature of its approach; but it is apt to
be used for total social action.
4. Finally, at the micro level, a similar but perhaps more subtle trap
can be seen; alas, I have been falling into it lately myself. For, just be-
cause of the imperfections of macro planning, a considerable number of "devel-
opment thinkers" have sought salvation in'micro-planning at the District; nay,
at the sub-District; nay, at the village level. For here we shall at last reach
reality the actual, complex conditions of motivation and of circumstance in
which the poor live; and here we can tailor action to full reality.
5. Quite complex, multi-disciplinary analysis is needed at this level.
In the rural area, an economist with farm-management training must be there;
surely a social anthropologist would be essential? The need for an agronomist
goes without saying; but what about a public administration expert with manage-
ment theory experience to suggest a system for running the Extension Service?
And a political scientist? for we cannot forget elitism, TANU, political
arenas and socio-political scenarios. An ecologist...?
6. To research-minded people this analysis is immensely attractive and
I (gladly) prophesy that a very great deal of it will be done over the next
ten years. But how does it look to the administrator and to the government?
Where are all these experts to be found? How much will they cost? How long
will they take to survey even a group of a dozen villages before anyone should
dare to suggest an executive programme which will meet all the difficulties
which their final report will enumerate?
7. I am sorry to ask these destructive questions, for I firmly believe
that the switch of attention to local realities, and the careful reconnaissance
of an area before a programme for it is imposed would represent a great advance,
not only from general ideological approaches but from central national direction.
But can we really believe that government in a large country will take on this
huge task of detailed diagnosis and detailed local prescription? Is it possible
to treat some millions of farmers in this nurse-maiding way?
8. What I am saying is NOT that the research should not be done, at least
in a far greater number of varied situations each of which might, to the best
feasible degree, typify an area. My thesis is really in two parts. First, that
the choice of action cannot simply arise from analysis (if sociological issues
arise, a sociologist must be sent in). Action is constrained by other consid-
erations by feasibility in manpower and expense; by knowledge of the capacity
of 3rd level field officers; and, above all, by the aim of providing, not a
thousand individual prescriptions but a physical and motivational environment
within which the main actors (farmers and officials) can use moderately en-
lightened common sense and discussion to find their own way through their part
of the forest. Government action must be broader and simpler than the sum of
9. The second part rests on a belief that sample researching at micro
level (more of it, and better done), will in fact reveal common factors, common
guidelines to conduct, by means of which commonsense can indeed be enlightened.
This, of course, implies a.belief that the local social process itself, enabled
by adequate infrastructural help, but not constrained by detailed regulation,
has the necessary dynamism to meet its own problems.
10. Ultimately, this means a belief in the free society.* But even rela-
tive simplicity of government action will not be achieved by.simple thinking,
or just saying "Participation," "Democracy." Just as clear and simple engineering
design is painfullyevolved from Heath Robinson complexity, so the lines of simpler
and more effective government action will need much hard and detailed work in
elucidating local facts and problems. The tabk of the Social Scientist remains,
to inform the designers of action; but the design will not follow their detailed
*Cf. "We all want to be God, and to manage things much better than God...by
eliminating just these possibilities of error in which human freedom consists."