The development of survey research is seen by many as one of the most important methodological breakthroughs in the social sciences. It has become the most widely used tool of empirical investigation in sociology and has found extensive and increasing use in disciplines ranging from anthropology to economics. For our purposes we will define a survey to mean systematic, comparable data on individuals obtained by means of one or more verbal methods of data collection, such as interviews or self-completed questionnaires. The popularity of this method lies in its ability to efficiently gather theoretically relevant data for enough cases (usually samples of larger universes) and in quantifiable forms so that statistical analysis can be carried out.

In the last decade, the use of surveys in social research in Africa has expanded considerably. During the colonial epoch, aside from an occasional cost-of-living study, an anthropological exploration into tribal life, or the often irregular census enumerations, surveys were relatively rare. The growth of universities with social science departments, the visits of foreign investigators, the studies required by outside lending agencies such as the World Bank, and the expansion of planning functions within government ministries have all increased the demand for surveys of various kinds. Researchers, supported by agencies like the United Nations, the Nordic Council and the Ford Foundation, have conducted investigations in collaboration with various governments as well as on their own. Their objectives have ranged from research into fairly abstract social phenomena, such as linguistic etymology, to measuring critical parameters that form the basic assumptions in planning operations, such as aspirations of school leavers.

Ever since African art was “discovered” as an aesthetic as opposed to ethnographical genre, enthusiasm for its emotional vigor and clarity of form has centered primarily upon sculpture. Sculpture, to be sure, is the most widespread mode of African artistic expression, and the description and analysis of regional and ethnic styles in sculpture either in the round or relief have commanded a great deal of attention; but, while carving in wood has been the chief medium modeling in clay has also been extensively developed. Some of the most interesting of African artistic achievements have been embodied in clay sculpture either in the round or in architectural reliefs, yet very little has been written on clay sculpture, and virtually nothing has been done on African architecture and its myriad patterns and styles of mural decoration. Reasons for this imbalance are not difficult to find: most of the books on African art have been written in Europe with the help of extensive museum and private collections. Mud sculpture, on the other hand, is extremely difficult to transport, especially if sun-dried and not fired; and architecture remains essentially impossible to study in absentia.

A few anthropologists have mentioned mud sculpture and architecture, but writers on African art have tended to ignore this area of both historic and contemporary artistic achievement.

The nations of Africa are today experiencing the cross-pressures characteristic of modern economic development. The choices of development strategies to focus these pressures constructively are critical, and no choices are more important than those which will determine the interactions between agricultural and manufacturing sectors.

Relationships between agriculture and industry have received wide attention in the literature of development economics. Generally the two sectors have been depicted as competitive for the scarce developmental resources available, with opinion then splitting over the relative emphasis to be placed on each at various stages of the development process. It will be argued here that three fundamental assumptions must be challenged if development strategies are to promote optimal economic growth. (1) The assumption that the fundamental nature of the agricultural-industrial relationship is competitive must be drastically modified. (2) Growth rates of aggregate output and output per capita can no longer be accepted as “the” indicators of developmental economic health. (3) Gross national (or domestic) product measurements must be rejected as providing the most relevant index of aggregate output. Only within a framework of reconstructed assumptions can the optimal relationship between agriculture and manufacturing be defined and achieved.

Structural transformation—“the process by which a predominantly agrarian economy evolves into a diversified industrial economy”—is generally recognized as an essential characteristic of economic development. Structural transformation encompasses many things. Agriculture's relative share in both output and employment declines, and increased specialization of economic functions within and between sectors greatly increases the interdependence of economic units.

The essay by Hopkins and Mitchell (“The Viability of Survey Research in Africa: Some Propositions”) appears to have multiple objectives which may benefit from further examination. First, it is an apologia for survey research in Africa which is intended to counter “doubts” and “attacks” which arise from some quarters. However, the criticisms of survey research in Africa need to be more fully identified along with evidence or arguments to discount them. In particular, the major criticisms cited in footnote 4 are not dealt with explicitly in the subsequent set of “propositions.” Thus the objective of defending survey research in Africa (or elsewhere) remains largely unfulfilled.

Second, Hopkins and Mitchell offer a catalog of “assorted conventional and not-so-conventional wisdom” on the design, execution, and evaluation of survey research in Africa. However, there are several potential problems with the propositions which should be noted: problems of coverage, generality and distinctiveness.

Coverage. As it now stands, the review is not sufficiently comprehensive in scope nor detailed in treatment to serve as more than a general guide for survey researchers. It is a useful beginning but merits considerable elaboration and systemization. Moreover, some of the propositions (e.g., 1, 4, 5, 16) appear to have little to do with survey methods perse but rather apply to social research in general.

In the nineteenth century one can distinguish between three different stages in the development of British military policy in West Africa. Up to 1840 British military policy relied heavily on eighteenth-century precedent and revolved round the triple pillars of the West Africa Squadron, the forts, and the military personnel, made up mainly of Europeans who were brought from Great Britain or raised among European volunteers at the coast in times of emergency. Between 1840 and 1870 the British relied rather heavily on the West Indian troops, whom they used in a strategic role to deal with military problems in her West African possessions. During this period attempts were made to raise small but local paramilitary units to supplement the efforts of the West India Regiments. From the 1870's onwards the limitations in the use of the West Indian troops and the need to cut down military expenditure led to the emphasis placed on the raising of comparatively large African and local forces which later took over the duties previously undertaken by the West India Regiments. Towards the end of the nineteenth century, when the need arose, these local forces were amalgamated into what came to be known as the “West African Frontier Force,” which eventually became the instrument used by the British government not only to ward off European trespassers in those areas which it claimed but also to consolidate its authority in West Africa in subsequent years.

The role of Algerian women in their own society has rarely been what it has seemed. Outside observers have often indulged in generalizations based on a lack of understanding, in judgments represented by only partial truths. David Gordon's recent pronouncement (1968) that “women are the serfs of Algerian society” is, for example, a fair enough description of the position of the majority, but it is, in the end, no more accurate than Jacques Berque's (1967) axion that “there were not even any bastards” produced by the temporary sexual relations between Algerians and European settlers in the 1930s. What is quite clear is that women have played and continue to play important roles in Algerian society although they have always suffered definite legal, social, and cultural disadvantages that have not been essentially changed by independence.

Historically, Algerian women have enjoyed more freedom than is usually admitted by Western authors. In rural areas women were allowed to go about daily routines without hiding. Only in cities was the wearing of the veil normal, and even this practice can be easily explained. Germaine Tillion's interesting and convincing interpretation (1964) on this issue comes to mind: the practice of wearing a veil in countries bordering the southern shores of the Mediterranean is simply an attempt of basically nomadic peoples to maintain privacy in the crowded cities. Nor is this custom unique; Spanish women wear black, a practice that serves much the same purpose as the veil in Muslim lands.

Professors Hopkins and Mitchell have asked me to comment upon the issues raised in their paper as they appear from the vantage of the discipline of anthropology. I should like to state at the outset that my own views are hardly orthodox. Like many others whose graduate education occurred in the 1960s, I took great liberties in ignoring disciplinary boundaries both in my education and in my research. I was, and still am, more interested in the processes of social and cultural change, of “modernization” and “development,” than I am in remaining an anthropological purist. The number of researchers interested in the study of social change continues to increase, and many of us find that a focus on these processes often orients us away from the traditional approaches of our disciplines and toward whatever methods and techniques prove to be useful for gathering information relevant to our studies.

I found in the course of my research in rural Tanzania in the late 1960s that extensive survey techniques complemented the intensive anthropological techniques of long-term participant observation and reliance on a small number of key informants. A growing number of scholars who identify themselves as anthropologists, or would be so identified by others—though not all of them—agree with such a position. Yet most would probably go so far as to say that while research based solely on in-depth, intensive techniques qualifies as anthropology, any research which is based only on survey techniques has lost many of the qualities which anthropologists and others recognize as characteristic of the discipline.

The literature on job satisfaction and work attitudes clearly shows that salary or money is one undisputed source of people's motivation to work. They work in anticipation for money which they get to exchange for most of their needs. Salary as a source of workers' dissatisfaction has teen revealed in several studies. Studies conducted in British factories (Marriott and Denerley 1955) and those carried out in the United States—one based on a national sample (Center and Cantril 1946) and another, a follow-up study of gifted children by Terman and Oden (1959)—all show that the income level and job satisfaction of workers are positively related. Reiner (1957) found that inadequate salary was one of the factors which made the teachers he studied leave the teaching profession. Research by Rudd and Wiseman (1962) on some 432 teachers from the University of Manchester School of Education revealed similar findings. Phipp's study (1968) of Ugandan teachers and Nelson's (1970) of Congolese teachers also revealed that salary was one of the dissatisfying factors indicated by the teachers.

Another facet of the role of money in the motivation of workers has been explored by Herzberg and his colleagues (1959). From the findings of their study, they have formulated their motivation-hygiene theory in which, using an analogy from medical field, they call those factors which they claim cause the dissatisfaction of workers—dissatisfiers—the hygiene factors.

If survey research is research that consists of interviewing large samples of persons, or asking them to complete a standardized questionnaire—and I agree that's what a survey is—then surely the social sciences would have long ago broken through to far more sophisticated methodologies for addressing theoretically relevant hypotheses. Interview and questionnaire studies are rather primitive data-collection techniques, however sophisticated the concomitant sampling techniques or data-analytic techniques might be.

Therefore, I find it a bit ironic that we are, in this symposium, finding so much to debate, and so many propositions to consider about survey research. What, after all, is there to say about survey research that hasn't already been said? Aha! you say, we're not talking about survey research per se, but survey research in Africa. Well, that is indeed so, but some of the most important things that must be said about survey research apply to survey research wherever it is done, in Dar es Salaam or Swarthmore or Syracuse. And before turning to some issues raised by the authors of the core paper in this symposium, which issues pertain particularly, if not uniquely, to survey research in Africa, let me make some general observations.

First, survey data are empirical data, and that's good, because empirical data are better than guesses, or deductions, or intuition, or even (I think) divine revelation. But survey data reveal only what people say in response to what the survey research asks.

All of us have very decided ideas about teaching. While our geographical interest may be Africa, our disciplinary training and specialties are varied, and our methodologies and ideologies lead us, not only into different research paths, but toward differing teaching goals and learning expectations. Perhaps, common interest in Africa creates a better understanding of, and cooperation among, the various disciplines, and, if that is the case, it is an indisputable advantage in our classes. Still, it would be presumptious, indeed impossible, for a historian to tell a geographer or an economist or a linguist how, or even what, to teach.

However, our profession expects two things of us: to produce original contributions to the pool of knowledge in our specialty, and to teach effectively what we have learned to undergraduates seeking a general education, as well as to graduate students whose educational goals are more akin to our own. There is no small irony in the fact that, for a variety of reasons, as the number of college students dwindles, as the cost of education spirals, and as the job opportunities for new Ph.D.s all but vanish, our peers are more demanding in their evaluation of our research, while college and university administrators are more concerned with “accountability;” that is, total student credit hours produced—our teaching.

There are three points I want to make about teaching, but first I think it is important to make a disclaimer.

I claim authority to provide an economist's view because (1) I am an economist and (2) I recently returned from Nairobi where I conducted an interview-survey of about 1500 African households, the data from which I am now processing with the aid of a grant from the Rockefeller Foundation. In the remaining five sections of this paper, I will discuss the value to Africans of survey research in Africa by expatriates, the value to expatriate scholars of such survey research, twenty-five survey-based studies in Nairobi, the Hopkins-Mitchell manual of survey research in Africa, and my conclusions.

In their paper, Hopkins and Mitchell state their purposes to be (1) to address the doubts raised by “many scholars and African officials about the value and appropriateness of surveys in Africa” and (2) “to stimulate discussion that will be of interest to prospective expatriate (mainly American) researchers in Africa.” The authors devote practically all of their discussion to advocating high quality survey research and to offering suggestions on how that research should be conducted. They barely address the other issue, namely, the value and appropriateness of surveys in Africa, and their brief treatment of it is clearly inadequate. After a brief but stimulating digression, I will return to discuss their argument.

Jungle imagery and prejudiced stereotypes still abound in American knowledge of Africa. In a 1967 survey of American schoolchildren, E. Perry Hicks and Barry K. Beyer found that the majority still saw Africa through a web of myth and unfavorable stereotypes which appeared to increase as students progressed from grade seven through grade twelve (1970, pp. 158-166). In closing remarks as President of the African Studies Association, Philip D. Curtin noted this dilemma and posed a compelling challenge to the Africanist community. “The 1970's,” Professor Curtin stated,

This essay is written in response to Professor Curtin's challenge and as an attempt to begin a dialogue on how to teach African history effectively. It is our contention that African studies is a recent enough arrival on the academic scene that teachers have not yet developed, or have not yet shared, techniques of instruction to cope with the special problems inherent in the subject matter.

Before analysing the success or failure of the single-party state in Africa, it is necessary to define the terms of success and failure of a political system. First, it must be emphasized that the prolonged existence of a regime does not prove its success nor that its overthrow is necessarily a result of its failure. Sometimes cruel, corrupt, and inefficient governments survive for decades while reform-minded, popular governments are overthrown within months. Franco's Spain, for example, has achieved little success in the realm of social and economic progress, although it has succeeded in surviving for more than thirty years. Dubcek's Czechoslovakia, although of short duration, is not necessarily a failure when viewed in historical perspective. If, then, the longevity of a regime has no bearing on its success or failure, what criteria can be employed in making an evaluation of one-party states in Africa? The three criteria which seem relevant for such an evaluation are those of stability, economic development, and nation-building. This paper, therefore, will be devoted to a comparison of the single-party state with Western-style multiparty systems and with military dictatorships , focusing on the three criteria mentioned above. At the same time, the paper will attempt to answer the question: does a single-party state succeed or fail in its performance because of its single-party structure?

An examination of politics and political thought in Tanzania since independence reveals three recurring themes. These are egalitarianism, mass participation in politics at all levels, and anti-elitism. A brief examination of these ideas will serve to point out some possible sources of political conflict if this ideology does become a guide for decision-making in Tanzania.

Equality implies the adoption of a universalistic legal order, the application of achievement criteria in recruitment in all sectors of society, and fair treatment for all citizens. However, like many other new states, Tanzania is culturally plural, is unevenly developed, and has skilled manpower shortages. In the drive to fill positions with nationals, it is doubtful that equality of opportunity can be achieved with any consistency in the immediate post-independence period.

The Arusha Declaration encourages mass participation in political life at all levels. Furthermore, compared to other African political parties, TANU has been relatively successful in involving citizens in the political process. The rationale for advocating mass involvement is to narrow the mass-elite gap and to create a sense of responsibility on the part of citizens. However, mass participation could also lead to the weakening of leadership by skilled individuals and the overloading of the political system via inflated demands.

The Arusha Declaration also deemphasizes the significance of wealth, education, skill, or other such values in the recruitment of leaders. The objective is to discourage elitism and all forms of elitist behavior on the part of the leadership.

In the fall of 1972 the International Labor Organization published Employment, Incomes and Equality: A Strategy for Increasing Productive Employment in Kenya. While this study has general implications for employment policies in underdeveloped countries, it is especially relevant for urban problems. We will here examine the extent to which the ILO's proposals for handling problems arising in the “informal sector,” within which hawkers and squatters are considered the most troublesome, are likely to be accepted by the Kenya government and the Nairobi City Council.

According to the ILO, the gap between the growth of the Kenya population (3.3 per cent annually) and wage-earning employment (less than 2 per cent) makes rising unemployment inevitable, particularly in the major towns, where the African population expanded during the 1960's at a rate of over 10 per cent per annum. To stem urban migration, or to expel nonwage earners from the towns, is impossible because of the scarcity of arable agricultural land. Likewise, the government cannot provide enough jobs for the unemployed without seriously overburdening the economy due to the country's inadequate industrial potential. Therefore, it is proposed by the ILO that the government should primarily concern itself with increasing employment opportunity by encouraging self-employment rather than wage earning.

The self-employed in Nairobi earn about $240 annually (L90) and provide real services for the low-income segment of the population—selling, exchanging, repairing, transporting, and even making or assembling much of what is needed by those unable to afford the amenities enjoyed by the elite.