Lifeworld and System: Of Water Systems and Grain Mill Development

Lifeworld and System: Of Water Systems and Grain Mill Development in Rural TanzaniaAuthor(s): Tony WatersSource: African Studies Review, Vol. 35, No. 2 (Sep., 1992), pp. 35-54Published by: African Studies AssociationStable URL: http://www.jstor.org/stable/524869 .

Accessed: 22/02/2014 17:01

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

African Studies Association is collaborating with JSTOR to digitize, preserve and extend access to AfricanStudies Review.

http://www.jstor.org

This content downloaded from 132.241.17.164 on Sat, 22 Feb 2014 17:01:17 PMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=afsta

http://www.jstor.org/stable/524869?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp


Lifeworld and System: Of Water Systems and Grain Mill Development in Rural Tanzania*

Tony Waters

I simply take it for granted that other men also exist in this my world, and indeed not only in a bodily manner like and among other objects, but rather as endowed with a consciousness that is essentially the same as mine... [and that] in the natural attitude of everyday life the following is taken for granted without question: (a) the corporeal existence of other men; (b) that these bodies are endowed with consciousness essentially similar to my own; (c) that the things in the outer world included in my environs and that of my fellow-men are the same for us and have fundamentally the same meaning; (d) that I can enter into interrelations and reciprocal actions with my fellow-men; (e) that I can make myself understood to them (which follows from the preceding assumptions); (f) that a stratified social and cultural world is historically pregiven as a frame of reference for me and my fellow-men, indeed in a manner as taken for granted as the 'natural world' (g) that therefore the situation in which I find myself at any moment is only to a small extent purely created by me (Schutz and Luckmann 1973, 4-5).

Rural development projects involve the application of bureaucratic systems of consciousness to traditional areas where there have never been such normative systems applied before. This means that the values, norms, and ideals of bureaucratic institutions are implicit to how traditionally oriented rural villages are approached by outside development agencies and also how relationships develop between them. If these two forms of consciousness-bureaucratic and traditional-are different they could have fundamentally different assumptions about the meanings of social exchange and reciprocal action. These problems are rooted in what Jurgen Habermas (1987) insists are the very different forms of rationality governing moral and legal action in traditionall villages and modern bureaucratic agencies. This article is about the problems that purposively-rational bureaucracies have working with villages grounded in traditional-rational ethics.

African Studies Review, Volume 35, Number 2 (September 1992), pp. 35-54.



AFRICAN STUDIES REVIEW

Implicit to any bureaucratic approach by a development agency is the ideal of purposive-rational social structures which sustain the agency (Habermas 1987, 209-12). Most, if not all, development assistance agencies have an implicit grounding in the assumptions of modernization theory, whether stated ideals are of the neoclassical or even of a dependency perspective. This is because such agencies are typically based on the utilitarian functions that agency subsystems must have in order to fulfill a specific function or purpose which is that agency's mandate. These rationalized subsystems are necessary in the broader systemic relationships between the individual development project and the rest of the modern bureaucratized world in which policy-making is done. Much of the rational-choice theory used to describe bureaucratic institutions in the modern world has taken advantage of this fact in order to describe relationships across a wide cross section of social structures.2

In differentiated systems such as bureaucracies, it can be presumed that a specific social action has a purpose which can be described in materialistic/utilitarian terms (e.g., "we are doing a because we need b"). This is true whether the purposive-rational theory be of the modernization schools, and emphasize markets and individual action, or the various world systems/dependency schools which emphasize the material interests of elites.3 The norms, ideals, and expectations of purposive-rational bureaucratic institutions, be they from Japan, Tanzania, or the United States, are endowed with consciousnesses which use similar moral and legal assumptions, despite the fact that the cultural bases may be dissimilar (Schutz and Luck- mann 1972, 4-5). These norms are based in utilitarian principles.

An example of how important such a similar consciousness is to economic change is illustrated by George Homans' description of how the English woolen trade emerged despite cultural differences. Homans points to how deeply embedded rational choice is in modern institutions: "The [utility theory] explanation depends on my assuming that there were a number of merchants, Flemish, English, and indeed others, who were concerned with the woolen-cloth trade and that, whatever their other differences, they shared substantially the same sort of values or at least many of them did...[but] I have certainly tacitly assumed that the actors from the king downwards, behaved according to their perceptions of what would be successful in increasing their [economic] rewards" (1987, 75; emphasis added).

But, what happens when there are not only differences in social values, but also in economic consciousness? What would have happened if the Flemish merchant or the Scottish shepherd had not defined the economic situation in the same way as the English king, or had traded not on the basis of utility, but, say, kinship relations? What happens

36



Lifeworld and System

when not only the bases are dissimilar, but the superstructure above the bases as well? I think the answer, as the example below illustrates, is that in such circumstances purposive-rational reasoning is not predictive of social action. Or, put back into Alfred Schutz and Thomas Luckmann's words, it is no longer possible to take for granted that "other men also exist in my world" (1972, 4-5).

As indicated above, this is what can happen when a development agency, which is organized along bureaucratically rationalized principles, approaches villagers embedded in traditional rationalities for social action. The presumed systemic articulation is lacking and, as a result, miscommunication is a likely consequence. The result can be project failure (i.e., in purposive-rational terms). In an undifferentiated traditional system, action may not have a single well-defined purpose, but is interpreted as having meaning throughout the entire social system. Habermas (1987) has written about such problems most recently, though, it should be noted that descriptions of such traditional consciousness are at least as old as Karl Marx's (1978, 608) complaints about the "homologous magnitudes" which were the French peasantry and, perhaps most notably, include Bronislaw Malinowski's descriptions of kula rings in the Trobriand Islands (1922), Marcel Mauss' classic essay The Gift: Forms and Functions of Exchange in Archaic So- ciety (1954), and Max Weber's descriptions of modern "purposive contracts" and traditional "fraternization contracts" (1978, 668-75). Robin Horton has described how traditional African rationality differs from the rationalized Western scientific thought which implicitly permeates our own theoretical thinking (1970). Peter Berger, Brigitte Berger, and Hansfried Kellner put such thinking in a broader perspective (1973). More recently, writers such as Stephen Bunker (1987), V. Y. Mudimbe (1988), Goran Hyden (1985), and David Lan (1985) have assessed how Western definitions for political and social action have been applied in Africa, both with success and without. Using the theoretical assumptions implicit to such a broad sweep of thought about traditional rationality, the following practical example illustrates the role that such differences in rationality can play in the implemen- tation of a development project.

Of Grain Mills and Waters Systems in Rural Tanzania

In remote areas of Western Tanzania where I worked from 1984 to 1987, village governments were able to organize effectively and operate village grain mills (Waters 1989a). This operation included the collection of user fees, importation of diesel fuel, procurement of spare engine parts, and technical maintenance. The village government which ran the grain mill was hierarchically organized using

37




purposive-rational models consistent with Tanzanian government policies. These village governments included a village chairman who was responsible to a village council, which was in turn assigned specific tasks by the village constitution. A party representative appointed from the district headquarters also frequently played an important role as an intermediary.4

Paradoxically, water pumps installed by the bureaucratized organizations of the Tanzanian government and foreign assistance programs remained idle, despite the fact that analogous benefits could be obtained from either technology. Ironically, these pumps were sometimes in the same villages which operated successful grain mills. Typically the water systems had been installed after agency personnel, consistent with agency, Tanzanian government, and United Nations policy, made contact with the village through the village hierarchy. Contracts with villages were subsequently approved by both the government or agency representative and a vote of the village council. Maintenance and operation, as with the grain mills, was the responsibility of the village governments.5

As a result of such policies during the period 1968-87 perhaps 30 percent of the villages had grain mills installed which operated at least sporadically, while 10 percent of the villages had water systems installed which had diesel pumps, none of which were operational. In order to further understand the broader conditions of such a paradox, at least in purposive-rational terms, the following notes are provided:

-Both the water system and the grain mill required reg- ular inputs of expensive imported diesel fuel and imported spare parts. Technically speaking, both had diesel engines which drove auxiliary equipment (i.e., pump or mill).

-The water systems replaced traditional sources of water, usually springs, which have traditionally been protected from pollution by village normative standards. Such normative standards also regulated the springs for fair use.

-The grain mills replaced hand-grinding which had been done on stones at individual homes. There were no analogous norms governing use of the grinding tools, which, unlike the water systems, were not inevitably shared.

-Labor saving benefits for both technologies accrued primarily to women who were relieved of daily chores, that is, grinding meal by hand in a mortar and pestle, or water carrying.

38




-Both technologies also appealed to aesthetic values. The mills produced a finer and tastier meal, while the water pumps produced a visibly cleaner and more de- sirable product. The water pumps also had long-term public health benefits which were not so readily apparent to the villager but were an important factor validating project funding with government and agency officials.

-Both technologies required substantial capital inputs for installation by the villagers. In the case of the grain mill, cash contributions are required for initial purchase of the machine, while for the water systems, substantial contributions in the form of unpaid self- help labor to dig trenches was required.

-Along with bicycles, grain mills and water pumps were the primary mechanical innovations introduced in this area during the last 20 years.

-Both the grain mills and water systems were typically operated by village governments consistent with the so- cialist ideals established by the ruling Chama cha Mapinduzi Party and central government. User fees were collected for each technology: for the grain mills in the form of direct user fees, while the unmetered water systems were, at least in theory, operated using village tax money.

-Kigoma Region was one of the last areas in Africa to come in contact with rationalized economic media. Un- til the turn of the century, all trade was conducted on a barter basis; the introduction of a purposive-rational economic media (i.e., money) was resisted violently by the Waha people until at least 1907 (Illife 1979, 118- 32).6 Illife reports that the introduction of money to the area in 1911 by German colonial authorities who wanted to facilitate tax collections was a basis for armed revolt.

These notes show that in purposive-rational choice terms, there is not any substantive reason why grain mills are used and water systems are not. But, the empirical observations remain. The grain mills were operational, and the water pumps idle. Did I (or the planners who at least implicitly implemented the "water first" policy consistently over 20 years) miss some factor which is part of the rational decision- making model used by the village? This, after all, is the type of

39




question which those of us working for purposive-rational bureaucracies are rewarded with promotions for answering correctly; during the last 20 years there must have been some incentive for planners, expatriate and Tanzanian, to explain what social fact caused the water pumps to be inoperative.

While in Kigoma, my colleagues and I speculated about the rea- sons for this anomaly. We asked Tanzanian government officials and villagers for their opinions as well. Initial comments blamed lack of material equipment such as diesel fuel or spares. After this argument was countered with the example of the grain mills, the following three explanations were typically offered, whether by expatriate, government official, or villager:

a) A feminist argument noted that the grain mills produced a tastier meal than hand grinding methods. Thus, the men who ran the village governments had in- centives to ensure that the grain mill continued to operate. The water pumps on the other hand provided a product which did not have such benefits for the men; the primary benefits accrued only to the powerless women-thus the lack of interest in maintenance and operation.

b) The grain mill could be run more like a business, with profits accruing (albeit illicitly) to the village leaders operating the mills, while public utilities such as water systems offered fewer opportunities for such manipulations. This is in effect a private enterprise argument.

c) Fee-for-service milling charges are perhaps more col- lectable than the monthly or annual tax collections necessary to maintain the water system consistent with Tanzanian law. This is basically a "free rider" argument: it assumes that the nature of the non-metered flat fee/tax charge to keep the communal taps running is more subject to abuse than the fee-for-service charge (i.e., a set charge for milling a certain volume of grain) typically used to finance the grain mills.

Although all three arguments do have elegant points to make, even in the context of a bureaucratically grounded rational-choice argument, none provided the key which would have illuminated the paradox, at least from a social fact perspective. Here is why.

The feminist argument (a) is too simplistic given the success of the grain mills. There is no evidence, at least in purposive-rational terms, that the men in Kigoma will first dig trenches for pipe, only to

40




turn around and withhold payment of service fees because the primary beneficiaries are women. If this argument was the key, the resistance and lack of cooperation would have been expected during the trench- digging projects, when the bulk of the capital investment was de- manded, and not after.

Likewise, with the private enterprise argument (b), a certain credulity is required. The assumption is that corruption is easier when operating a grain mill than with a water system. In fact, the opposite may be true: an efficiently run water system (i.e., one operating daily) requires larger inputs of valuable diesel fuel and other materials with proportionately larger potential for skimming, price fixing, and kick- backs potentially associated with either operation.

Of the three purposive-rational arguments discussed here, the "free rider" argument (c) is often found to be the best, even though it is still lacking in some ways. Certainly, unmetered water taps are subject to abuse and inequality. With any such utility, someone will always live closer to a tap and there is always potential for illicit use of the community resource. However, this also is true for the traditional water sources in the area, usually springs which are also used communally and therefore subject to free rider abuses. And the springs have been successfully regulated by village norms for cleanliness and fair use for many years.

Thus far, this article has advanced three different explanations as to why water pumps are inoperative, while an analogous technology, grain mills, are successfully operated. As has been shown, none of these explanations provide a convincing explanation. In the sections that follow, an alternative explanation of the program is proposed which relies on Habermas' differentiation between lifeworld and system in hopes that the problem can be rephrased in a manner which can offer more explanation. This is done by recognizing that up to here, this paper has outlined two related paradoxical questions:

1) Why are grain mills maintained and operated effectively while water systems are not, that is, why is the "grain mill first policy" implicitly adopted; and

2) Why do villagers embedded in traditional-rational structures offer purposive-rational accounts of social action, even though they are presumably embedded in traditional-rational social structures?

This paper up to here has demonstrated that purposive-rational approaches provide an inadequate means of addressing these questions. But, this raises the question of whether there are any other, more ap- propriate, approaches to issues of rural third world development, and if so, what they are. After all, purposive-rational approaches do

41




explain a great deal, if imperfectly, and cannot be easily dismissed in the absence of an alternative.

Habermas' (1987) model of lifeworld and system provides both a basis (if not a specific answer) for why grain mills worked while water systems did not and a more theoretical assessment of why villagers offer purposive-rational accounts of their activities even though they may be embedded in traditional-rational systems of reasoning. Habermas' model can provide a framework for a purposive-rational assessment of traditional-rational actions. To show how this can be done, I will first present Habermas' model within the context of what is generally known about village values in Tanzania, and then analyze the two paradoxes presented above.

System and Lifeworld

In Habermas' model (1987), the force driving modernization is the differentiation of a traditionally rational "lifeworld" (Schutz and Luckmann 1973) and the emergence of a purposive-rational "system." In the traditionally oriented villages such as those described here, the rationality, or basis for communicative action, is a traditional one which focuses on kinship and relations between segmentary kin-groups in their economic and social functions.7 Social interaction emerges out of a lifeworldly morality which focuses on preserving given kin and/or lineage relationships. In the lifeworld of such traditional villages, decisions are made by consensus, and thus, even when systemic requirements-such as grain mill operation-are necessary, it is a need which emerges out of lifeworldly arrangements based in village values rather than purposive-rational decisions based on an analysis of costs and benefits. Such forms of consensus8 are in turn based in validative accounts which emphasize magic or kin-based relations. These are the concepts which Horton (1970) describes as being the "theoretical" basis for African society, meaning the set of "taken for granteds" which in our society are grounded in the functionally equivalent theories of science. Based on Habermas' description, then, it can be postulated that the reason that there was no effective purposive-rational explanation for the grain mill/water system paradox is that the "grain mill first" policy arose out of lifeworldly considerations which are not purposive rational, but embedded in non-purposive criteria which have their roots in a traditional episteme which presumably emphasizes kin- relations.

This is hardly a new observation. The substantial literature about traditional villages in Africa and elsewhere indicates that social interaction emerges out of a lifeworldly morality which focuses on preserving given kin and/or lineage relationships.9 This means that

42




the actual key or social fact based in purposive-rational terms might be irrelevant or, in modern purposive terms, irrational. While presumably such kinship arrangements may be evaluated on a utilitarian, cost- benefit basis, this is an awkward proposition at best as any number of project proposals which have justified a utilitarian decision in a non- utilitarian social structure evidence. But the alternative, that is, making a traditional-rational justification to a large purposive-rational bureaucracy, is a contradiction in terms. Imagine writing a report to the highly purposively-rational United Nations that this money was ill- spent because it disrupted how one group of spirits felt about another to pay cash for water, but not grain milling, so the water system did not work, despite obvious utilitarian benefits.

Obviously, profit in such traditional circumstances, at least in terms of rational-choice theory, has very different meanings than that presumed by the purposive-rational assumptions of development assistance bureaucracies.10 Thus, even when specific systemic requirements (such as grain mill operation) are necessary, in Habermas' formulation these are needs which emerge out of lifeworldly arrangements based on consensus and group advantage, rather than contracts between representative individuals (e.g., village council members and agency staff).

Does development of economic and social institutions, though, require an organically-or in Habermas' terms systemically- functional social structure? In Habermas' description of modernity, social action is uncoupled from lifeworld contexts, as purposive-rational attitudes are developed towards social/contractual ties and principles of validity based on consensus are replaced with principles of individual rights and majority rule.11 Money and power enter the equation and enable the calculation of individual strategic advantage in ways which were not possible before. Such calculations enable social actors to bypass traditional consensus-building mechanisms as they seek strategic advantage not only as kin-groups, but as individuals (Habermas 1987, 183; see also Weber 1978, 674). With this limitation of the lifeworld, "the lifeworld contexts in which processes of reaching understanding are always embedded [i.e., consensus] are devalued in favor of media-steered [i.e., money] interactions; the life-world is no longer needed for coordination of action" (Habermas 1987, 183).

However, while the lifeworld is no longer needed for such coordination, it does remain in modem society as a differentiated subsystem of the larger purposively-rational system (Habermas 1987, 173). As a result, the larger purposive-rational structure remains anchored in the lifeworld, where unquestionable concepts of validity, such as morality and truth, continue to be found; indeed "[e]very new leading mechanism of system differentiation must...be anchored in the lifeworld; it must be institutionalized there via family status, the authority of office, or

43




bourgeois private law...that can come about only if the lifeworld is suf- ficiently rationalized, above all only if the law and morality have reached a corresponding stage of development" (Habermas 1987, 173). As the purposive-rational differentiation of subsystems occurs during modernization processes, the nature of the lifeworld also must shift.

This means that in terms of communication there are very different criteria for validity in differentiated modern society than in undifferentiated traditional society. It is very different from the model for traditional society because, in large part, the organizing principle of social relationships, that is, the contract, excludes lifeworldly validity claims, first by explicit conflict of interest regulations, and secondly by including clauses in written contracts which nullify verbal commitment. An important indicator of how important such contractual relationships are to differentiating modern from traditional society is provided by Weber (1978, 666-81). Like Habermas, Weber stresses that such "purposive contracts" are critical to the ordering of modern society. They supplant what Weber calls traditional "fraternization contracts" in which economic exchange symbolizes a broad lifeworldly relationship such as father, wife, master, slave, comrade, protector, and client. To "'fraternize' with another person did not, however, mean that a certain performance of the contract [in purposive-rational terms] contributing to the attainment of some specific object, was reciprocally guaranteed or expected" (Weber 1978, 672). And here lies the crux of why water systems in Kigoma do not work while grain mills do. Simply put, the water systems are embedded in two separate but contradictory contractual systems (modern and traditional), while grain mills with their origin in the traditional lifeworld do not have such a contradictory legal basis; their roots are completely in traditional society.

In traditional systems functionally similar validity claims, or "taken for granteds" (Habermas 1987, 124), are based on a consensus and have roots not only within a differentiated and rationalized lifeworld, but throughout society. Unlike the differentiated system of the purposive-rational contract, such validity claims are implicit to discourse and embedded in the legal/moral systems which govern normative behavior throughout society. Again, reference to the classical work of Mauss (1954) and Malinowski (1922) but also more recent work by Mudimbe (1988), Berger, Berger, and Kellner (1973), and Ranger (1983) are probably the best examples for demonstrating how economic considerations can and do permeate lifeworlds of traditional societies without reference to defined subsystems. In such systems "conflicts of interest" do not make sense, nor do clauses about the non- binding nature of verbal commitments. Indeed, such restrictions, if they were somehow enforceable, would bring social activity to a halt.

44




In modern purposive-rational systems, though, this lifeworld is in a differentiated subsystem as are sectors such as grain mills or water systems. Contracting for a rationalized physical structure such as a water system is in a separate subsystem which is carefully separated from other subsystems including kinship relationships or political relations. Claims for validity are explicitly stated in contracts which are also explicitly signed and sealed. The only lifeworldly "taken for granteds" left are the meanings implicit to the purposive-rational contract itself. For, while the contract itself is explicit, there are still taken for granteds, including 1) differentiated purposive-rational action, 2) nonconsensus based agency policy which steers field negotiators, and 3) calculations of cost-benefit ratios based in a concept of a differentiated project using media-steered mechanisms (i.e., money and statistics) which are irrelevant to traditional values and morality.

Locating Water Pumps in System and Lifeworld

Returning to the grain mill/water system paradox, it can be shown how assuming that two explicitly different decision-making processes were present can begin to explain the paradox. The agencies were operating with the assumptions that: 1) the villagers had purposive-rational political criteria and had selected a village council which had a systemically based responsibility to make decisions about the village infrastructure; 2) such decisions were made in a separate subsystem which was not influenced by other subsystems; and 3) this organically grounded village council articulated well with the field staff of the agency to whom, likewise, was delegated the systemic authority to make decisions based on systemically generated agency policies which indicated that water systems (and not grain mills) were an important priority. In the case of water systems, such purposive- rational reasoning often is grounded in a concern for health and sanitation standards.

The villagers were at the same time making an agreement which they believed was grounded in traditional forms of mechanical solidarity. Assuming that this was happening, this meant that equipment (water, pipes, etc.,) were accepted in a spirit of negotiation between two social entities, for example, agency and village/kinship group. Pipes and pumps were exchanged as symbols of emerging alignments, and labor to dig trenches was offered in return, but also as a symbol of the emerging relationships (Malinowski 1922; Hyden 1983; Bunker 1985; Weber 1978, 668-75). Since the exchange was unequal, there was probably an assumption of clientage proffered; this patron-client relationship disappeared when the agency withdrew or, in purposive- rational agency terms, "handed over" the system. In traditional terms

45




it was simply an end to the clientage exchanges and thus concern with the water system. Thus, there is a notably different relationship between the grain mills, which presumably have an origin in the village lifeworld, and the water systems, which have an origin in a contractual patron-client relationships.

In terms of the agency's lifeworldly "taken for granteds" the chairman and village council have an authoritative function which can articulate with the function of the authorized agency personnel. A village constitution is in modern terms a contract between council and villagers and has definite meanings about power and authority within the situational horizon as defined by purposive-rational logic.

In contrast, the same situation occupies a very different role from the perspective of the villager. From an undifferentiated lifeworldly perspective, there are two amorphous (unsegmented) entities dealing with each other-first the village or kinship unit, and second the agency--each of which is pursuing its own, perhaps complementary consensus-based group interest. Ties are presumed to be on a strictly lifeworldly level, which means that items given-in the case of water systems this means pipes and pumps-have a significance symbolic of emerging relationships (Mauss 1954), rather than sub-systemic purposive rationalized uses intended by the donors (e.g., the water is intended to improve health standards). This relationship, given the obvious economic inequalities between agency and donor, typically emerges as the traditionally based patron-client relationship embedded in what Hyden (1985) describes as the "economy of affection" which is in turn embedded in a resilient segmentary "peasant mode of production" (see also Habermas 1987, 123, 219-22).

System and Lifeworld-Grain Mills

The inconsistencies between traditional-rational and purposive- rational would seem to make the future of development cooperation rather dismal; with such inconsistency, there seem to be few mechanisms through which change can occur. However, the empirical evidence does not support such a conclusion, not even with this example. After all, grain mills with their roots firmly in the purposive-rational world of diesel fuel procurement, technical maintenance, and individual fee collection do work in Tanzanian villages. This, at least, is an indication that some materialistically oriented decision making has been made and that some tasks are delegated in purposive-rational manners. After all, fuel for grain mills is not brought to remote villages by consensus, and, particularly, neither is money collected on a fee for service basis to run the machine. The rational-choice paradox described is between a choice of technologies (grain mill or water system) rather

46




than the apparently more straightforward purposively rational operation of modem technology.

Horton has something to say about this when he writes "a person seeking to place some event in a wider causal context has a choice of theories...this choice will depend on just how wide a context he wishes to bring into consideration" (1970, 143). What is indicated is that in Tanzanian society there are different sources of rationality known to villagers. This is particularly so since the purposive-rational action implicit to modern government began making its presence felt in colonial times, but particularly in the post-independence era when the new government ambitiously reached out into remoter areas. This leads up to paradox 2, which asked how villagers embedded in traditional- rational structures can offer purposive-rational accounts for social action.

The use of purposive-rational social facts as validative accounts (i.e., paradox 2 above) is an important indicator of shifts in self- perception, identity, and the legitimacy of validative accounts. There is an implicit recognition by persons grounded in traditional-rational society that other views, at least tentatively, exist. The contrast between the violent resistance to the purposive-rational attempts to introduce money in 1907 and the introduction of water systems in 1968 is instructive; resistance as such has shifted to an entirely different plane. Such shifts in resistance would seem to point very clearly toward a further differentiation of lifeworld and system; remember, the accounts offered (the feminist critique, larceny hypothesis, and free-rider argument) use purposive-rational accounts to validate traditionally- based values, rather than the other way around.12

It seems that the changing validity of purposive rational reasoning with consequent delegitimation of traditional-rational values in public spheres becomes a discreditable part of the social identity for the individual (e.g., the village chairman) and the society as an en- tity. As such, traditional-rational reasoning becomes a carefully managed part of the identity, particularly when out-group alignments are established (see Goffman 1986, for a general description, and Waters 1989b, for a specific description from Kigoma14). Such careful management has implications for how future society will develop in traditionally private spheres of action (e.g., within the village), which must eventually be reconciled with the publicly presented identity necessitated by contact with the world system. This ultimately is the type of mechanism through which the newly differentiated system re-anchors itself in the lifeworld. Ultimately, how identity is managed with out-groups has implications for new lifeworldly criteria which emerge socially and culturally.14

47




Of Theory and Development

Theoretical approaches to development have been handicapped by reliance on purposive-rational type reasoning. Implicit to such reasoning are value-laden concepts about progress, development, and modernity. As has been pointed out elsewhere (Berger, Berger and Kellner 1973), the first two terms in particular imply values grounded in technological and materialistic norms. By dint of association, such norms have at various times become associated in a functional manner with other characteristics of Westernization, including development of bureaucratic institutions, industrialization, market economies, the nation-state, and the host of attitudes and beliefs associated with such institutions. The examples of the grain mills show that the association between a technocratic operation and modern purposive-rational norms is not as close as might be presumed. This is particularly evident in the context of the water systems failures.

Such failures cannot be understood by social theories grounded in concepts of linear development. Nevertheless, the bulk of the development literature suffers from such approaches whether it be of the more traditional modernization theory still used by international aid organizations or macro-approaches of the world system/dependency school which emphasize center-periphery relations. The emerging in- formal economy literature, of which Tanzania has begun to produce fine examples (Hyden 1981, 1985; Maliyamkono and Bagachwa 1989), does touch more on the issues raised in this paper, though without necessarily grounding description in any theoretical understanding of social action.

To a certain extent, the grounding of such ideas in Western thought is perhaps inevitable; after all, this paper, this journal, and the readers of this journal inherently have a grounding in the lifeworldly norms of Western discourse (Mudimbe 1988; Chatterjee 1986). However, the inherent limitations of such a process do not mean that our models and understanding of processes in Africa cannot be further re- fined.

Thus, it is the purpose of this paper to put one practical development situation into such a theoretical framework. Admittedly, substantial assumptions and generalizations have been made about the nature of society in Kigoma. But the assumptions presented here require shorter leaps of faith that those already used by planners necessarily projecting their understandings and consciousness about development onto the rural villagers concerned.

48




Conclusions

Changes do occur as traditional-rational societies adapt to purposive-rational approaches during modernization. The question is how does this happen, and, while it does happen, how can the agents of purposive-rational bureaucracies communicate with traditional- rational societies having feet in both worlds? As this example demon- strates, intuitively familiar pre-interpreted realities sometimes overlap in ways significant enough to permit communication between the two worlds.

An important observation to be drawn from this discussion is that legitimized purposive-rational behavior enters traditional society at different times. Thus, purposive-rational decisions can be made about the operation of a grain mill, even though what in a fully developed modern society is a cost-benefit decision (i.e., the choice of a grain mill rather than water system) may still be embedded in the undifferentiated lifeworld. This would seem to be an important indication of how differentiation between system and lifeworld proceeds.

Any model of communicative action which is to be of use in ana- lyzing development issues needs to take account of these differences. This means that existing purposive-rational project development needs to embed itself more in traditional-rational reasoning if it is to be more effective. Is this possible or is it a contradiction in terms? Apparently not completely, since it must have not infrequently occurred at least "subconsciously." And, judging from what has been written previously about successful development projects (see in particular Fowler 1982; Hyden 1981, 1985, 164-87; Waters 1989c), it must have happened at least in the subconscious collective of development agencies, no matter how hegemonic the purposive-rational ideal may be. Particularly in smaller agencies much of development management is done on a personalistic basis, at least at systemically unofficial or casual levels. Indeed, the corruption that is so often decried in development projects is often the result of personalism of a traditional-rational sort gone awry, rather than faults attributable to a more purposive-rational sort of kleptocracy occasionally found in the U.S. savings and loan industry. Hyden has perhaps written most directly about problems with nepo- tism associated with traditional-rational "economies of affection" in Tanzania (1981, 1985).

Indeed, tolerance of traditional-rational attitudes can serve very practical utilitarian purposes in project development (Fowler 1982; Waters 1989c). The ultimate measure of the purposive-rational is not the principle involved, but whether or not in utilitarian terms, the job was done. At least in Western terms, the acceptance of such utilitarian terms as validation for social action, for better and worse, is ultimately

49




what development, modernization, and the differentiation of system and lifeworld have been all about.

Notes

This paper has evolved out of many conversations with a number of people who lived in Kasulu area from 1984 to 1987. Conversations with Fr. Hank VanderParver W. F. of Kakonko, and Karin Mahn of Kasulu were especially influential. Mzee T. B. Siweya and Mzee N. P. Mchonchele also contributed to the ideas described here. John R. Hall at UC Davis pointed me in the theoretical direction that this paper has taken. Much to his (Weberian) dismay, he pointed out that the absence of a Durkheimian social fact is what makes this paradox interesting. Professor Hall also assisted me by discussing Habermas' ideas with me at length. Two anonymous refer- ees who read an earlier version of this paper also pointed me in directions that I had not thought of previously. All of these people as well as my wife Dagmar have my thanks and appreciation for what they have directly and indirectly contributed to my thinking, though I am solely responsible for the contents of this paper.

1. The adjective traditional is used in this paper with an awareness that grouping of pre-industrial societies is a perilous enterprise because it does not reflect the diver- sity of cultural forms in the traditional world. However, it is also readily apparent from the long-established ethnographic literature that there is a major discontinuity in pre- and post-capitalist social formations. This was observed at least as early as Marx, but in the classical sociological literature is perhaps described most elegantly by Durkheim in his essay about mechanical and organic solidarity. In modern Tanzania, the most elegant description of such traditional forms has beer: written by Hyden (1981, 1985) who describes the pre-capitalist peasant modes of production which are embedded in the "economy of affection." The use of the word traditional in this paper is meant to be a shorthand for the complex of values and world views described by Hyden, but which ultimately emerged from the writings of the earlier sociological classics. I am aware that there is often an implied linear evolutionary relationship implied between traditional and modern, but know of no good way to avoid this implication. I use the term with a reluctance similar to Berger, Berger, and Kellner (1973), but also like to emphasize that the dichotomy is necessary to describe very real historical periods in which material, social, and technological forms are very different. Unlike Kuper (1988), I do not be- lieve that such a dichotomy reflects simply a need for the West to identify and classify the other. Rather, it reflects the very real, if hegemonic, influences that modern technology, bureaucracies, and economic organizations have had in shaping the world.

2. See for example Hechter (1987). For a description of modernization theory, see Ros- tow (1959).

3. See for example Cardoso and Faletto (1979), and Wallerstein (1976). For a critique of how world systems theory is applied to Africa, see Mudimbe (1988, 3-23).

4. Hyden (1990) and von Freyhold (1979) both describe well the official Tanzanian system of village government. It is a system which has local village elections, but the selection of candidates is controlled by central authorities of both the government and ruling Chama cha Mapinduzi Party. In theory, the elected village leaders are the link between the masses and the central government which appoints officials to administer government programs. Moves between centralization and decentraliza- tion since independence have had varying effects on relations between the govern-

50




ment and village. However, as a rule, a great deal of authority has been retained over appointments by the central government in Dar Es Salaam.

5. The first question asked by a development agency approaching a village is the classic "take me to your leader." This is both good manners and very practical, at least from the purposive-rational position of the agency and the government. In modemn Africa, this generally results in a contact with the government or party representative in the village who, it is assumed, has a hierarchic role relative to the villagers. Agencies interpret this role as being functionally similar to the purposive- rational roles found in their own hierarchies. As a result, if the leader fails to de- liver on his hierarchical responsibilities, it is assumed that this is because another person in the village is in fact the real functional leader. Development agency dogma has it that such a person could perhaps be a pastor, shaman, woman's leader, or elder, in the same way that a secretary or other bureaucratic function is likely to control a bureaucracy from behind the scenes. Such an interpretation though still assumes purposive-rational structures and, as a result, misses an important point about the nature of undifferentiated society. In fact, what is the problem is that in an undifferentiated society, there is no leader in purposive-rational terms.

6. The difficulties that the Germans had with incorporating the Waha into the colonial system are also noted in Iliffe (1969, 15). The Admiralty War Staff (1916) of Great Britain described the Waha living north of the Malagarasi, (i.e., the area described here) as "imperfectly conquered" by the Germans at the time World War I began in German East Africa. For older descriptions of Kigoma, see in particular Scherer (1959). Grant (1925), Leakey and Rounce (1933), and Tawney (1944) wrote ar- ticles about Kigoma from their perspectives as colonial officers. More recently, Gwassa and Mbwiliza (1976) and Brain (1973) have offered scholarly accounts about social relations in the region.

8. Or in Durkheim's term, mechanical, and in Marx's term, homologous. 9. Consensus and unanimity can continue to be achieved in traditional villages in ways

no longer known in the purposive-rational West. Hyden recently observed that local leaders prefer to be elected unanimously, which he believes is an indication that traditional norms, rather than democratic norms, about consensus-building continue to prevail in Tanzanian village politics (1990, 302). Significantly, I think, this is a different interpretation than such an election would generate for the purposive- rational (and democratic) West. In purposive-rational societies, such unanimity is an indication not of consensus, but of authoritarianism. Of course, there is a fine line between the two phenomena, and care needs to be taken in interpretation.

10. See for example Malinowski (1922), Mauss (1954), Fortes and Evans-Pritchard (1940), Hyden (1981, 1985), Horton (1970), and Habermas (1987, 666-81). For examples from Africa, see Mudimbe (1988), Hyden (1983), Lan (1985), and Slater (1976). For a specific description form Kigoma, see Scherer (1959).

11. Or, for that matter, the African governments which assumed authority from depart- ing colonials assumed much of their purposive-rational ideals. See Ranger (1983) and Hyden (1981).

12. See also Hyden (1990). 13 What I have been calling "traditional rationalization" is not inherently inconsis-

tent with participation in the modern world, even if the means to do so are not so easily defined in the purposive-rational exercise which is this paper. Lan (1985) in his descriptions of the role that spirit mediums played in legitimating the Zimbab- wean Civil War among peasant populations makes quite clear that social customs and beliefs are not necessarily embedded in a differentiated lifeworld. He claims

51




that such social customs were able to successfully challenge the Rhodesian state. If what Lan says is true, (i.e., that spirit mediums and traditonal belief continue to play a role in ordering social action in the Zimbabwean state) it is reasonable to ex- pect an analogous situation in traditional Tanzanian villages.

14. Waters (1989b) describes a circumstance where traditional norms resulted in decreas- ing the number of children claimed for each household in a census tied to a disaster aid program. In this case, despite the utilitarian (i.e., purposive-rational) advan- tages of making a truthful, or even exaggerated, estimate for the number of children in a household, the aid recipients systematically underestimated children in defer- ence to a local taboo which was of a magical nature. This is a good example of how private, traditional-based values embedded in magic are presented in out-group relations. The magic was still respected even at the expense of material needs, while still presenting a modern purposive-rational identity to the out-group. Notably, Horton has noted that such beliefs are common in Africa where to know the name (or in this case even to acknowledge) a thing or person is to have power over it (1970, 155-56).

15. By the same token, this gap is frequently crossed by field-level development workers who subconsciously adopt a personalistic management style which can reflect traditional-rational norms rather than those of the bureaucracy. Workers who do this to too great of an extent are considered to have "gone native" by the purposive- rational agency employing them, with the result that further promotion within the bureaucratic hierarchy is blocked. Scherer (1959) encountered similar attitudes during his study in Kigoma in 1952/3.

References

Admiralty War Staff, Intelligence Division. 1916. A Handbook of German East Africa. Great Britain: Naval Intelligence Division.

Berger, Peter L., Brigitte Berger, and Hansfried Kellner. 1973. The Homeless Mind. New York: Random House

Brain. James. L. 1973. "The Tutsi and the Ha: A Study in Integration." Journal of Asian and African Studies 8: 39-49.

Bunker, Stephen. 1987. Peasants against the State: The Politics of Market Control in

Bugisa, Uganda 1900-83. Urbana: University of Illinois Press. Cardoso, F. and E. Faletto. 1979. Dependency and Development in Latin America. Berke-

ley: University of California Press. Chatterjee, P. 1988. Nationalist Thought and the Colonial World: A Derivative Dis-

course? Totowa, NJ: Zed Books. Fortes, Meyer, and E. E. Evans-Pritchard, eds. 1940. African Political Systems. London: H.

Milford. Fowler, Alan. 1982. "Temperatures in Development Funding: The Hot Money Model." De-

velopment (Journal of Society for International Development): 3. Goffman, Erving. 1987. Stigma: Notes on the Management of a Spoiled Identity. Engle-

wood Cliffs: Prentice-Hall. Grant, C. H. B. 1925. "Uha in Tanganyika Territory." The Geographical Journal 66: 411-

22. Gwasa, G. C. K. and J. F. Mbwiliza. 1976. "Social Production, Symbolism and Ritual in

Buha: 1750-1900." Tanzania Notes and Records 79/80: 13-21. Habermas, Jurgen. 1987. The Theory of Communicative Action, vol. 2. Boston: Beacon

Press.

52




Hechter, Michael. 1987. A Theory of Group Solidarity, Berkeley. Berkeley: University of California Press.

Homans, George. 1987. "Behaviorism and After." In Social Theory Today, edited by An- thony Giddens and Jonathan Turner. Cambridge: Polity Press.

Horton, Robin. 1970. "African Traditional Thought and Western Science." In Rationality, edited by Bryan R. Wilson, 131-71. Oxford: Basil Blackwell.

Hyden, Goran. 1981. Ujamaa in Tanzania. Berkeley: University of California Press. . 1985. No Shortcuts to Progress, African Development in Perspective. Berkeley:

University of California Press. . 1990. "Responses from Below, A Tale of Two Tanzanian Villages." Food Policy

August: 299-305. Illife, John. 1969. Tanganyika Under German Rule 1905-1912. Cambridge: Cambridge

University Press. . 1979. A Modern History of Tanganyika. Cambridge: Cambridge University Press.

Kuper, Adam. 1988. The Invention of Primitive Society. London: Routledge. Lan, David. 1985. Guns and Rain: Guerillas and Spirit Mediums in Zimbabwe. London:

James Currey Ltd. Leakey, E. A. and N. V. Rounce. 1933. "The Human Geography of the Kasulu District,

Tanganyika, The Land of the Abaha." Geography 18: 293-305. Malinowski, Bronislaw. 1922. Argonauts of the Western Pacific: An Account of Native En-

terprise and Adventure in the Archipelagoes of Melanesian New Guinea. New York: Dutton.

Maliyamkono, T. L., and M. S. D. Bagachwa. 1989. The Second Economy in Tanzania. Athens: Ohio University Press.

Marx, Karl. 1978. The Marx-Engels Reader, edited by Robert C. Tucker. New York: W.W. Norton.

Mauss, Marcel. 1954. The Gift: Forms and Function of Exchange in Archaic Society. Glen- coe IL: Free Press.

Mudimbe, V. Y. 1988. The Invention of Africa, Gnosis, Philosophy and the Order of Knowledge. Bloomington: Indiana University Press.

Ranger, T. 0. 1983. "The Invention of Traditions in Colonial Africa." In The Invention of Tradition, edited by Eric Hobsbawm and T. O. Ranger, 211-62. Cambridge: Cambridge University Press.

_ 1985. Peasant Consciousness and Guerilla War in Zimbabwe. London: James Currey.

Rostow, W. W. 1959. "The Stages of Economic Growth." Economic History Review. 12/1: 1-16.

Scherer, J. H. 1959. "The Ha of Tanganyika." Anthropos. 54: 841-904. Schutz, Alfred, and Thomas Luckmann. 1973. The Structures of the Lifeworld. Evanston:

Northwestern University Press. Slater, Mariam. 1976. African Odyssey. Bloomington: Indiana University Press. Tawney, J. J. 1944. "Ugabire A Feudal Custom Among the Waha." Tanganyika Notes and

Records 17: 6-9. von Freyhold, Michaela. 1979. Ujamaa Villages in Tanzania: Analysis of a Social Exper-

iment. New York: Monthly Review Press. Wallerstein, Immanuel. 1976. The Modern World System, vol. 1. New York: Academic

Press. Waters, Tony. 1989a. "Of Water Systems and Grain Mills: Observations of Felt Needs in

Western Tanzania." ISSUE: A Journal of Opinion 18/1: 47-50. _ 1989b. "Some Practical Notes on a Names Taboo in Western Tanzania." Disasters 13/2: 186-87. . 1989c. "Money and Development: Necessities and Contradictions." Vita News

July: 17-19.

53




. 1992. "A Cultural Analysis of Economies of Affection and the Uncaptured Peas- antry in Tanzania. Journal of Modern African Studies 30/1: in press.

Weber, Max. 1978. Economy and Society, Volume 2. Berkeley: University of California Press.

54



Lifeworld and System: Of Water Systems and Grain Mill Development

Documents