Atelier 8, article 3


© The democratisation of decision making processes in the water sector

© "Journal of Hydroinformatics" www.iwapublishing.com

Fichier au format PDF : ici

Part 1

M.B. Abbott
International Institute for Infrastructural, Hydraulic, and Environmental Engineering, Delft, The Netherlands, and Knowledge Engineering, BVBA, Brussels, Belgium

ABSTRACT : The paper introduces an inversion of the structure of the decision making process that has been so far followed in most countries, and in almost all so-called ‘third world’ societies. It is commonly observed that the general population becomes alienated and effectively disempowered through the existing ‘top-down’ knowledge transmission process. The disenfranchisement of large parts of the general population and the grievous harm to these parts through their resulting disempowerment has led to an outcry against the water professionals, who are seen at the very least as accomplices, and often as initiators in ‘crimes against humanity’. Empowering the population as a whole as genuine stakeholders in water resources then becomes the basic objective of water professionals in introducing an alternative paradigm as exemplified in the second part of this paper by the design of a new system capable of supporting ‘knowledge-intensive’ agricultural practices.

KEYWORDS : democratisation, decision-making, sociotechnical system, hydroinformatics

1 THE CRIMINALISATION OF THE WATER PROFESSIONS

An increasing number of popular publications, ranging from newspaper articles to complete books, are sowing the notion among the general public that water professionals are at least collaborators, and in many cases instigators, of actions that do so much harm to such large numbers of persons that they might be castigated as ‘crimes against humanity'. At the same time, the legal basis for the notion of ‘crimes against humanity' has itself become more firmly established by a number of bodies, ranging in turn from international organisations to national law-giving assemblies. These efforts have led to the formation of such bodies as the International Courts of Justice, situated in The Hague, the Netherlands, which have already sentenced a number of persons to long-terms of imprisonment for ‘crimes against humanity' involving direct physical violence.
Alongside this stream of development there is another, in which groups of citizens come together to oppose the actions of certain existing institutions and organisations that are perceived by these groups as unethical to the point of behaving criminally. The most dramatic (because most widely publicised) of these oppositions occurred in Seattle in November 1999, when a combination of organisations of citizens were able to prevent the holding of the regular WTO conference and its surrounding activities. However, other such protests, less publicised and usually less successful in meeting their objectives, have long taken place in the water sector. Going back ten years, for example (Roy, 1999, pp.46-48):

On Christmas Day in 1990, six thousand men and women walked over a hundred kilometres, carrying their provisions and their bedding, accompanying a seven-member sacrificial squad that had resolved to lay down its lives for the river. They were stopped at Ferkuwa on the Gujarat border by battalions of armed police and crowds of people from the city of Baroda, many of whom were hired, some of whom perhaps genuinely believed that the Sardar Sarovar was ‘Gujarat's life-line'. It was a telling confrontation. Middle Class Urban India vs. a Rural, predominantly Advasi, Army. The marching people demanded they be allowed to cross the border and walk to the dam site. The police refused them passage. To stress their commitment to non-violence, each villager had his or her hands bound together. One by one, they defied the battalions of police. They were beaten, arrested and dragged into waiting trucks in which they were driven off and dumped some miles away, in the wilderness. They just walked back and began all over again. 

The struggle in the Narmada valley lives, despite the State.

It is important to realise the relative scales and essential natures of the activities which are involved in the water sector as compared with those that are presented as ‘crimes against humanity' elsewhere. Thus (ibid, p.19):

That's what it works out to, thirty-three million people. Displaced by Big Dams alone in the last fifty years. What about those who have been displaced by the thousands of other Development Projects? At a private lecture, N.C. Saxena, Secretary to the Planning Commission, said he thought the number was in the region of 50 million (of whom 40 million were displaced by dams)..

It may be safely assumed that the reduction of life expectation alone among these displaced and consequentially destitute persons amounts to the likes of some millions of complete human lives. Moreover (ibid, p.21):

A huge percentage of the displaced are Adivasis (57.6 percent in the case of the Sardar Sarovar dam). Include Dalits and the figure becomes obscene. According to the Commissioner for Scheduled Castes and Tribes it's about 60 percent. If you consider that Adivasis account for only 8 per cent and Dalits another 15 per cent of India's population, it opens up a whole other dimension to the story. The ethnic ‘otherness' of their victims takes some of the pressure off the National Builders. It's like having an expense account. Someone else pays the bills. People from another country. Another world. India's poorest people are subsidising the lifestyles of her richest.

Thus these millions of one racial group are being physically displaced, impoverished, and in a large measure eliminated by persons of another racial group, the last process of which has the precise name of genocide. However, ‘crimes against humanity' that can be construed as genocide are already the subject of prosecutions and sentences at the International Courts of Justice, and then for cases involving ‘only' the elimination of two or three thousand persons, and then of persons not so different genetically from their eliminators but differentiated more by religious beliefs.
It seems clear that just to the extent that water professionals come to be perceived as accomplices in ‘crimes against humanity' and especially when they appear as instigators of such ‘crimes', they must in time expect to be arraigned before the appropriate courts of law. This will be the natural and inevitable consequence of the presently ongoing process of criminalisation of the water professions.

2 THE RESPONSE OF THE WATER PROFESSIONALS

Of course, among the water professionals themselves there will be few if any who will subscribe to this view. For by far the greater part, they see themselves as having carried out their professional duties to the best of their abilities. If any fault at all is to be found, they will say, it must be sought elsewhere, such as at the political and administrative levels of government. Some may perhaps accept that they may have been beguiled into activities with unfortunate consequences, but even these would see themselves, at worst, as what Lenin, an expert in the manipulation of well-meaning persons, described as his "useful idiots". 
Despite these denials and disclaimers, however, over the last decades, at least, there has been a gathering realisation among some water professionals that many problems of the water sector in many societies, and especially in societies in the so-called ‘third world', are themselves to a greater or lesser degree consequences of the actions of these professionals. There has in fact been a marked increase in the preparedness of water professionals not only to speak out the truth about their various failings, but, just as important, to listen to the truth. Thus, for example, when talking about "Learning from the past" in the area of capacity building in the water sector (Alaerts et al, 1999, p.18):

"The World Bank has been instrumental in putting capacity on the African development agenda. Nevertheless, an evaluation in 1994 indicated that only 28% of the projects achieved their objectives in terms of strengthening institutions, and only 23% were likely to produce sustainable capacity building benefits. The reasons include: (a) the Bank's rather narrow vision of capacity building, (b) a project cycle that inhibits capacity building, (c) an inordinate reliance on technical assistance that precludes capacity building, (d) badly formulated civil service reforms, and (e) the neglect of higher education in capacity building. The Bank may even have contributed to capacity erosion. It has tended to substitute for Africa's capacity deficiencies, rather than build capacity. Despite the rhetoric of self-reliance, all too often governments have found cosy accommodation with dependency."

At the same time, as things currently stand it is not normally the water professionals themselves who initiate changes that might improve the current situation. Thus, to take the situation prevailing in India (Alaerts et al, 1999, p.21):

"Already since the eighties, sector professionals called for delegation of decision-making power to the lowest appropriate level. For water supply and irrigation, this would mean setting up of local utilities and water user associations.
But in India, as in many other countries, the local government level (Panchayat) was not recognised as an executive entity by the Constitution. Hence, no Panchayat could formally take part in decision-making processes, or in maintenance of infrastructure. It could not even open a bank account.
The 1992 Constitutional Amendment gave this power, and thus created vast new opportunities for better water provision.
However, it was not the water sector's plea that caused this fundamental change, but rather external forces, that this time benefited the water sector ….."

Despite developments of this kind, failures not only persist, but appear even to accumulate in the water area. In his own analysis of World Bank assessments of project failures, Alaerts concluded that (Alaerts et al, 1999, p.52):

"Many failures can be attributed to systemic deficiencies in the institutions that determine policy, project design and operational management."

Moreover, the overall policy that underlies these failures remains basically unchanged beneath the surface of the Bank’s new-found, ‘progressive’, rhetoric. Thus (Fidler, 2000):

"Figures compiled by the Bank for a report suggest that 223 of its projects under way last October [i.e. 1999] would lead to the involuntary resettlement of 681,000 households and more than 2.6 million people… An earlier draft on the same report dated in May [1999] said 40 % of all projects involving resettlement were ‘likely to have significant adverse environmental impacts’…almost all having a ‘potential adverse impact’. However, these breakdowns were excised from the final version of the report."

3 PERCEIVED FAILURES OF ENGINEERING INTERVENTIONS IN THE THIRD WORLD WATER SECTOR

Starting some thirty years ago, to the author's knowledge, with criticisms of irrigation projects for promoting cotton production in certain Asian states of the former Soviet Union, the volume of condemnation of engineering interventions in the water sector has grown exponentially. As exemplified above, from being a concern of only a few ‘ outside experts’, this rising tide of discontent, which is now moving in some places towards a torrent of outrage, has caught up many millions of persons who have been directly influenced by such engineering interventions and has generated an increasing volume of ever more serious and detailed criticism in support of these persons. The point is more and more often made that not only have such engineering interventions as dam constructions and irrigation schemes failed to deliver many of their expected benefits, but they have had many negative consequences, and for the most part for the poorest parts of the populations concerned and for the natural environment. We may group together some of the principle criticisms as follows:

3.1 The special property to water, that it moves itself under no other influence than an always present and gratuitous gravity, makes it very easy to transfer the benefits that water confers from one community to another with little or no physical expense. Thus one region can enrich itself at the expense of another, either by laying legal claim to the waters of this other region, or by damming up or otherwise extracting the flow for its own use and thus denying it to the other region.

3.2 In this process of redistributing the benefits of water, the overall economic considerations are often swept to one side. Thus, going back to an earlier period, initiated in the USA of the 1930s (Reisner, 1993, p.119):

"Economics mattered little, if nothing at all; if the irrigation ventures slid into an ocean of debt, the huge hydroelectric dams authorised within the same river basin could generate the necessary revenues to bail them out (or so it was thought).
.... The natural landscape of the American West, was to undergo a man-made transformation the likes of which no desert civilization has ever seen."

3.3 Such projects commonly deliver very different results to one part of the population to the results that they provide to other parts: their benefits are often very unequally distributed even across the society that receives the water. This is very clear in the case of dam constructions, which have led to the formation of large lakes that in turn have necessitated the displacement of many people. As observed above, it has been estimated that the construction of some 3,300 large dams in India since independence has led to the displacement of between 30 and 40 million persons, for by far the larger part wrenched from their livelihoods with no compensation. (Only those owning land and satisfying certain citizenship conditions could claim some compensation as Project Affected Persons [PAPs].) Irrigation schemes have similarly led to relatively small groups that benefit and larger groups that are disadvantaged, and often severely disadvantaged, by such schemes. Increases in public health risks are often clearly pronounced in many of these schemes and these tend to affect the poor far more than the better situated inhabitants.

3.4 The so-called ‘secondary physical effects’ of such projects are often severe and effectively irreversible: a much augmented evaporation of water from dam-enclosed reservoirs and serious sediment erosion and accretion problems arising from dam constructions, and the effects of waterlogging and salination in irrigation schemes, are well-known examples. Remedial actions are usually difficult, if not politically impossible, and often prohibitively expensive.

3.5 The protection of land by empoldering in some areas commonly leads to the raising of water levels and longer durations of flooding in areas upstream of the empoldered region. There is generally again a shift of damages and associated costs from one part of the population to another part.

3.6 The construction of infrastructural projects together with water-related projects again leads to new shifts in employment, business activity and wealth, such as towards intensified truck-based traffic operations at the cost of ferry-boat operations. The net social effect may well be negative.
3.7 The large scale of most projects is often associated with monotonic agriculture the sustainability of which depends on large inputs of agrochemicals and mechanisation, both of which necessitate increases in working expenses, imports and associated costs. They also have deleterious consequences for groundwater and, especially in the case of herbiculture and insecticides, on the ecosystem generally. These developments are reinforced by the introduction of genetically modified (GM) crop species that may necessitate new seed purchases at each new planting, reduce ecosystem diversity and thus endanger ecosystem stability. For example (Shiva, 2000):

"Spending on pesticides in Warangal, Andra Pradesh, has shot up from $2.5 million in the 1980s to $50 million."

3.8 The construction of new projects often disrupts the existing methods of cultivation of the land and water, it breaks up long-established cultural patterns, it disrupts the ritualistic framework of social cohesion and otherwise causes grievous social harm. The destruction of existing agriculture practices together with the social, and especially cultural, contexts with which these are synergic is seen increasingly as a major threat both to the peoples of the third world and their natural environment. For example (Shiva, 2000):

"Recently I was visiting Bhatinda in Punjab because of an epidemic of farmers suicides. Punjab used to be the most prosperous agricultural region in India. Today every farmer is in debt and despair. Vast stretches of land have become water-logged desert. And as an old farmer pointed out, even the trees have stopped bearing fruit because heavy use of pesticides have killed the pollinators – the bees and the butterflies…

For me it is now time radically to re-evaluate what we are doing. For what we are doing to the poor is brutal and unforgivable. This is especially evident in India as we witness the unfolding disasters of globalisation, especially in food and agriculture."

Alongside this, on the side of the natural environment, the consequences are now seen to be nothing but catastrophic. The latest report of the World Conservation Fund shows that, for example, nearly one in four of the world’s mammal species is threatened with extinction as a result of human interventions, most of it in agriculture (Houlder, 2000). 
3.9 The relatively limited number of beneficiaries of these projects is not normally sufficient to pay back the capital, with interest, that has been used to realise the projects. Instead the debt burden has to be passed on to the population as a whole, a large part of which has however been so disadvantaged by just these projects as to be unable to contribute to the necessary repayments. In the words of the Director of the Centre for International Studies at Harvard University (Sachs, 2000):

"The IMF and the World bank have been mouthpieces of deception [of poverty alleviation] with their charade of ‘debt sustainability’ of the poorest countries. These analyses have nothing to do with debt sustainability in any real sense, since they ignore the needless deaths of millions of people for want of access to basic medicines and nutrition… The IMF knows very little about economic development challenges, from disease to tropical environment to environmental degradation…The World bank is equally ineffectual."

Similarly, speaking for Oxfam about the IMF-World Bank ‘debt relief’ programme for highly indebted poor countries, Watkins (2000) observed that:

"failure to reduce debt to [truly] sustainable levels will have grave consequences for social sector financing… Fiddling the figures while debt continues to destroy the lives of desperately poor people is not an appropriate response."

3.10 The construction of projects in this way serves the interest of certain influential political constituencies and these reinforce the power of the state officials, including engineers, who elaborate and supervise projects. The added presence of so-called ‘first-world’ consulting engineers and contractors, working in collaboration with local consultants and contractors, facilitates the availability of loans for such projects from the large lending agencies. The major sources of scientific knowledge which have been implanted into ‘third-world’ societies, such as the Surface Water Modelling Centre and the Environmental GIS Centre in Dhaka, can themselves only become sustainable business operations by supplying knowledge to the consultants and contractors which are engaged in this kind of activity. Knowledge passed to the state apparatus usually goes nowhere at all. Thus, for example, information supplied to the Public Health Offices or the Agricultural Extension Offices in Bangladesh may be most valuable in itself, but it is the essential nature of these centres that introduces the basic problem. Being governmental institutions they carry the burden of bureaucracy, but they are at the stage at which they have few suitable communication devices, and, most essentially, they appear to have no real commitment to provide the services to the general population for which they were originally created in the first place. To reverse this within the existing paradigm appears to be exceedingly difficult, if not impossible.
Within the present dominant paradigm an apparently financially-sustainable, technical-administrative ‘caste’ is created and maintained, but this is almost entirely self-serving, providing few services to the great mass of the population as a whole and practically nothing at all to the poorest. This situation is so obviously shocking that even the carefully-selected review panels and other such missions that assess various aid projects can hardly avoid missing it, and these regularly make recommendations to improve the situation. The consequences can never be more than cosmetic, however, since the entire structure that has been put in place in this way is inherently inimical to such interests of the general population. Indeed it is already quite widely accepted that the knowledge and experience of organisations with ‘grass-roots’ foundations are essential when assessing what are the real needs of people and what is the extent of existing knowledge when considering other kinds of services than those which are directly water related. This is especially accepted in the case of advice in the field of agriculture, but it also applies to the advice related to marketing products, access to wider markets, market opportunities for crop diversification and similar opportunities. This knowledge comes primarily from microbanking and other such organisations that have their own kinds of ‘philanthropic vested interests' in the alleviation of poverty, but it may also originate from other existing organisations, such as many NGOs working in different areas. However, in the initial stages of the ‘new paradigm' of knowledge provision that is introduced shortly, the input about the kind of knowledge that has to be provided, usually in the form of advice, has to come for much the greater part from people involved in ‘grass-roots’ networks, simply because they are so much closer to the principal ‘sources' and ‘end-users’ of this knowledge.

4 THE QUESTION CONCERNING EMPOWERMENT

It should be emphasised that water professionals are in a very special, and possibly even unique, situation as compared with all other professionals. Water is essential to all life on Earth and to the human economy as well, and the danger confronting life on Earth and the human economy is greater in the case of water sector than it is anywhere else. 
The realisation of the systemic nature of the existing and now well established structure of knowledge/power in the water sector has then naturally led to questionings from several persons and at various levels in the existing hierarchy. In some cases this questioning has led only to the outright rejection of the existing structure. One manifestation of this is the call by many professionals linked to ‘third-world' societies to turn a collective back upon the whole panoply of modern technology and science and to return to ‘traditional’ farming and other practices. Essentially, the same reaction is apparent in the case of the established knowledge centres in many ‘third-world' countries, where an exceptionally high proportion of younger staff leave every year for more advanced studies and possibly future employment in the ‘first world’, in many cases outside of the water sector. In other cases, however, this situation has led to a search for another way of making use of science and technology that would make their fruits available to the greater part of the population, and on a much more equitable basis. Of this latter approach, which is that which is followed here, we may observe the following:

4.1 It is this search that has led to the identification of an alternative approach which can make use of certain facilities that have been built up in ‘third-world' societies This approach is then essentially anti-systemic: even though it proposes to use certain facilities which have been established and developed within the existing paradigm, it proposes to use those facilities in entirely other ways than those for which they were originally intended, as well as for entirely different, and indeed opposite, purposes.
4.2 The essential feature of the change in paradigm that is thereby announced is its reliance upon an already accomplished change of paradigm, such as already been realised by the micro-banking community in some countries, and principally in Bangladesh. The basic strategy is to introduce a new structure in knowledge / power relations in the water sector in these countries, and indeed a structure that is anti-systemic with the existing structure, preferably by allying itself with organisations that are already anti-systemic with respect to the more generally established social paradigm in banking, mobile telephony and other areas that are closely related to the water sector. At a deeper level, which will be identified in the second part of this paper, we have to do here with engineering a change in knowledge/power relations by introducing changes in knowledge/value relations.
4.3 The pattern of the taking into possession of water rights by some at the cost of the dispossession of others of their water rights, as realised through the application of technology, has to be broken. This can only be done by the application of another kind of technology which is in turn embedded in another kind of social context.
4.4 As is now being increasingly accepted elsewhere, the mobilisation of people and the dissemination of the right information in appropriate form is more critical than the provision of new ‘hardware'. The definitions of ‘the right information’ and ‘appropriate forms’ are the central concern of the water-technical persons involved here, and these definitions lead to the posing of totally-integrated social, technical and scientific problems. The mobilisation of people and their provision with the right kind of information appears in particular as a central concern of the micro-banking and other organisations directed to raising the living standards of the general population, including the poorest.
4.5 As hydroinformaticians we naturally ask how we, as professionals, can change this situation for the better. As sociotechnologists we then ask in the first place why some protests, such as the WTO protests in Seattle, succeeded, and why the protests in India for the most part failed. Anyone who was unaware of the reasons before September 2000 could no longer be aware of the reasons for success in the one case and failure in the other after the protests that swept across Western Europe against increases in taxation occasioned by rising oil prices. Thus, from an editorial in the Financial Times (September 16, 2000):
A week ago, British politicians and investors could look at the lorry blockade in France with some satisfaction. Industrial strife [in Britain] was a thing of the past.

…he extraordinarily effective protest by hauliers and farmers has destroyed this illusion…

…conomic links are not limited to physical transportation. For all their dislike of their continental competitors, British truckers and farmers are in a similar boat and have learnt from the success across the English Channel. Modern communications enabled them to be highly effective without needing backing from a union or industry federation. The government found it almost impossible to counter such unstructured protests effectively."

In this case, ‘modern communications’ does not refer only to television broadcasts of large numbers of ‘ordinary’ people - farmers, lorry drivers, business owners - arguing their case with passion and conviction, and expressing their feeling of being disenfranchised within the present political system, but it refers also, and much more significantly on the organisational side, to the use of internet and mobile telephones. As mobile telephony moves further into broad-band communications with the third-generation capabilities promised by the year 2003, thereby offering all internet capabilities, so the possibilities to organise and coordinate such protests will grow immensely.

The lessons for the water professionals even in the first world must be quite clear: the general public must be empowered in a responsible and equitable way or otherwise this public will empower itself, and possibly less responsibly and very likely, through the influence of special-interest groups, in inequitable ways. Once again the technical side of the sociotechnical system required to do this must be provided by the new generation of telecommunication equipment and services. This mode of operation has already led some water professionals to investigate how the general public can be so empowered with responsible knowledge/value that this general public will itself behave in a responsible way. Now, however, the water professionals have to proceed yet further again, to provide tools that will inculcate attitudes of social responsibility and consideration, and this not only towards human societies but also and inseparably towards the world of nature.

5 THE IMMEDIATE PURPOSE AND NATURE OF THE NEW APPROACH

The immediate purpose of the developments introduced here is to provide the peoples of the ‘third-world', including the very poorest, with knowledge, initially in the form of advice, about all water related risks and opportunities with which they may be confronted at any time. Within this perspective, water is seen as a great unifying agent, within human and natural economies and, ultimately, between these economies—as has been celebrated over the millennia in any number of myths, legends and sagas of the peoples of the ‘third world'- and as such continues to be sustained by ritualistic practices. Within the framework of this paradigm, it is such autochthonic forms of knowledge, such as are grounded in a collective wisdom, that are regarded as the foundations of all other forms of knowledge of by far the greater part of the populations of these societies. Accordingly, by far the greater part of the knowledge and the means of providing advice are already implicitly present among the ‘ordinary’ people, including, if only potentially, among the poorest. The manner in which this knowledge and corresponding advice is expressed is comprised of those combinations of ordinary language, simple illustrations and practical emulation that we shall call, quite generally, ‘narrative discourse’.

(Now in much postmodern writing, and especially in the area of literary criticism, the word ‘narrative’ has acquired a rather specific connotation and this has more recently entered into the discussion of the preferred structure of broadband entertainment. Given media that lend themselves, and indeed are directed, to interaction between the ‘entertainer’ and the ‘audience’, in which these two roles may interchange between the participants, the notion of ‘narrative’ has become associated with (modern, rather than postmodern) forms of entertainment, such as are best suited to broadcasting. The latter are associated with ‘linear programming’ in which ‘stories’, understood as content that has a definite beginning, a definite middle and a definite end, unfold in time. This is to say that they have a broadcaster-given ‘story line’. It is then argued that broadband entertainment will be better served through the provision of means to produce non-linear content, as a kind of ‘emergent’, self-structuring content. Such a change in direction naturally involves very profound changes in attitude, and especially within the existing media companies. The aim of a system of the type to be described in the second part of this paper is to provide interactive, non-linear, content for some purposes, but not for others, with the constraint that the non-linear content must also run on more traditional media as well as third-generation (3G) telecommunications equipment. While fully accepting the relevance of these issues, the word ‘discourse’ is still however used here in its more traditional, mundane, sense, as any recounting of experiences that are either facts or appear as facts, so that it applies here to non-linear content provision, as well as linear.)

This kind of autochtonic knowledge and its practical everyday extensions, together with the ‘narrative discourse’ that embodies it as ‘narrative knowledge’, has more recently been joined in most parts of the world, and to a very considerable degree in several nominally ‘third world’ Asian and Latin-American countries, by another kind of knowledge, which we call ‘scientific knowledge’. Inseparable from this is another kind of discourse, which we call ‘scientific discourse’. By the application of scientific knowledge through the agency of scientific discourse it has become possible to do many things that were much more difficult, and even impossible, using the ‘traditional’ forms of narrative knowledge and narrative discourse. Thus, for example, in the present context it now becomes possible to foresee weather, flood and soil conditions considerably further forward in time, to estimate crop yields subject to various agricultural practices accordingly, and so provide advice on preferable farming operations. There are, of course, very many such examples of the possible benefits of applications of modern science and technology. At the present time, on the other hand, few of these kinds of benefits of scientific knowledge are available for most of the people of the ‘third world', and very few indeed are available for the poorest of these peoples.

One of the objectives of the present approach is to reinforce the existing pool of narrative knowledge with all available scientific knowledge and to make the resulting knowledge available for all the people of the ‘third world', including the poorest, by couching this knowledge in narrative forms that these peoples can assimilate and thereby integrate with their existing, personal and autochtonic, knowledge. Clearly, however, the value of this service will still be quite limited unless it is customised to interact with the existing knowledge and interests of its individual end-users. This is because the nature of the knowledge required as well as its mode of presentation will often vary greatly according to the ‘traditional’ knowledge and general situation of the individual recipients of this knowledge: the state of their physical and knowledge assets, their obligations, capacities, intentions and every other factor of this kind. The value of the integrated body of collective narrative knowledge/wisdom and scientific knowledge, no matter how appropriately this is presented technically, will be limited unless the full nature of each individual recipient ‘on the ground’ can be identified. This identification and the means of such identification must then also be translated from its existing narrative form into a form that can interact with the other forms of collective knowledge. 

The other immediate aim of the new paradigm is then to direct the attention of all available narrative and scientific knowledge to the real-world constraints, opportunities and aspirations of the individual recipients of that knowledge. From this point of view there is no such thing as a ‘rice farmer’ per se, but a great variety of different individuals, families and communities who cultivate one or more different varieties of rice together with various other crops and exhibit a very wide range of physical and knowledge situations, social situations and levels of ambition. This is to say, to introduce the technically precise term, that the repositories of science and technology will have to be provided with intentions that mirror the needs of the recipients of their knowledge if this knowledge is to be truly relevant. These repositories of knowledge have thus to seek out rather precisely the knowledge that is most relevant to each individual recipient of this knowledge and translate this knowledge into the best advice for each such individual recipient, at the same time providing the means to formulate this knowledge in an appropriate narrative form. By the same token, these repositories of knowledge have themselves to be provided with knowledge, in appropriate (and for the moment, in practice) scientific form, about the various situations and intentions of the individual families and communities that are the recipients of the narratively-presented scientific knowledge. 

The requirement that the repositories of the knowledge have to be structured and ordered in modern-scientific and correspondingly potentially technocratic forms may understandably appear as unfortunate to many persons, and especially the most well-meaning. After all, we can identify within modern science the very sources of the spiritual narrowness and subsequent wrong-mindedness of the existing technocratic paradigm in this field (Abbott, 1999b, 2000b). However, the originators of the new paradigm cannot themselves entirely escape the culture of modern science and technocratic attitudes with which they have themselves been formed (Husserl, 1948//1973; Abbott, 2000b). Even the vocabulary and rhetoric required to give expression to this proposed new paradigm cannot escape from this exigency. We thus have to do here inevitably with a two-way flow of knowledge, with one way proceeding from the ‘ordinary people’ to what are unavoidably for the moment the ‘scientific people’ and the other way proceeding from the ‘scientific people’ to the ‘ordinary people’. Both directions of flow have to pass through a ‘translation layer’, in the one case corresponding to a translation from narrative discourse into scientific discourse and in the other case corresponding to a translation from scientific discourse into narrative discourse. In the case of applications to agriculture of this paradigm we may then speak of a knowledge intensive agriculture. In much the same way we could just as well speak of a knowledge intensive aquaculture, a knowledge intensive health provision and other such developments. The pivotal and irreplaceable role of hydroinformatics in these initiatives is discussed later in this paper.

There is a further object of the new paradigm that is by no means so immediately and explicitly expressible, which is to maintain, and indeed in many cases to re-establish, a more sustainable balance between the needs of the human and the natural economies. We should observe, on the one side, that many of the traditional myths, legends and sagas in all civilisations are concerned with just this balance, while many traditional agricultural and other practices provide means to maintain the balance. On the other side, the joining together of narrative knowledge and scientific knowledge in the way that is now being proposed provides new and powerful means to overcome the increasingly common situation whereby, the more that mankind encroaches upon the needs of nature, the more nature responds by releasing ‘natural catastrophes’ upon mankind.

6 THE BASIC STRUCTURE OF THE CORRESPONDING SOCIOTECHNICAL SYSTEMS

The new paradigm proposed here introduces an inversion of the structure of the decision making process that has been so far followed almost everywhere in the ‘third world'. The structure employed almost exclusively so far is schematised in Fig.1.

It is commonly observed that the general population becomes alienated and effectively disempowered through this process. Empowering this population as genuine stakeholders in water resources, as schematised in Fig. 2, is the basic objective in designing the structure of a new class of systems.

What is being proposed here as a new paradigm thus corresponds to an inversion of the so-far established order. It corresponds to an inversion in power relations that is realised by an inversion in knowledge relations.

The systems that are proposed here as means of realising this inversion, which are essentially knowledge management systems, can clearly be organised in many ways: many architectures present themselves for this purpose. For the introductory, explanatory, purposes of this paper, a quite closely-coupled and concentric architecture is described, but it is well understood that more loosely coupled and more distributed, or more general ‘agent-orientated’, architectures may well prove to be preferable in practice. In the architecture introduced here, then, the principal components in the resulting system are as follows: 

The corresponding structure can be visualised as shown in Fig. 3.

This is clearly a hydroinformatics system. It is correspondingly within the present context a sociotechnical system: its technical components cannot function effectively without the simultaneous introduction of appropriate social (including institutional) arrangements and the social changes that it should release cannot be catalysed, and thus cannot be sustained, without the provision of appropriate technical means. These socio-institutional components are related as also shown in Fig. 3 (see also Abbott and Jonoski, 1998; Jonoski and Abbott, 1998; Thein and Abbott, 1998; Abbott, 1999a; Yan et al 1999; Thein and Abbott, 2000; Jonoski 2000). 

In basic outline, such a system is composed of:

1.A core group of professionals initially responsible for the design and analysis of the system and specifically of its centre, its inner periphery and its outer periphery.
2.The resulting knowledge centre.
3.The inner periphery (with other kinds of ‘professionals’) and:
4.The outer periphery (with a very few ‘professionals’ in the conventional sense, and mainly run by otherwise ‘lay persons’, and predominantly because necessarily by women, and then to some extent also necessarily on a part-time basis. (See Abbott, 2000).
The core group is responsible for the first realisation of the system and the establishing of its components. It comprises, so to say, the ‘construction team’ of the system. 

The knowledge centre is the primary repository of knowledge rendered in the form of modern-scientific and current technological discourse. It has:

a) to draw upon knowledge and data emanating from the outer periphery, as modulated by the inner periphery; and: 
b) to communicate with a variety of (other) knowledge providers; and:
c) to provide the outer periphery with appropriate narrative and scientific knowledge, originally mostly with a scientific form of expression, but arriving at the outer periphery in narrative form, having been mediated by the inner periphery.
The inner periphery has:
a) to ‘translate’ knowledge about the individual end-user arriving from the outer periphery that is primarily expressed in narrative discourse into knowledge for the centre that is primarily expressed in the manner of scientific discourse; and, on the basis of this knowledge;
b) to ‘translate’ knowledge and data arriving from the centre that is primarily expressed in the manner of scientific discourse into advice departing to the individual users situated within the outer periphery that is primarily in the form of an existing knowledge-sympathetic narrative discourse; and:
c) to process data exchanges between the centre and the outer periphery. 
The outer periphery is the primary repository of narrative knowledge. It has:
a) to inform the centre, through the intermediary of the inner periphery, what is happening ‘on the ground’ and especially the situations of each of the individual end users, and, further to this again, to communicate what these end users expect to happen (on the basis of an often long experience and their own individual knowledge); and: 
b) to draw upon scientific knowledge emanating from the centre, as merged with autochtonic knowledge by the centre, and as translated into narrative form by the inner periphery in order to provide advice to individual end users in the local population; and:
c) to act as a knowledge conductor between individuals and communities within the outer periphery.
REFERENCES

Abbott, M.B., 1999a, Introducing hydroinformatics, Journal of hydroinformatics, 01.1, pp. 3-19,.

Abbott, M.B., 1999b, Forchheimer and Schoklitsch: a postmodern retrospection. Proceedings of XXVIII IAHR Congress, Graz, Austria.

Abbott, M.B., 2000a, The gender issue in hydroinformatics, or Orpheus in the Underworld, Journal of hydroinformatics, 02.2, pp. 87-104.

Abbott, M. B. 2000b, Fides and Ratio, a critical appreciation, Journal of hydroinformatics, 02.2, http://www.iwaponline.com/jh/002/jh00200e1.htm.

Abbott, M.B. and Jonoski, A., 1998, Promoting collaborative decision-making through electronic networking, Hydroinformatics ’98, Babovic & Larsen (eds), Balkema, Rotterdam.

Alaerts, G.J., Hartvelt, F.J.A. and Patroni, F.-M., 1999, Water Sector Capacity Building: Concepts and Instruments, Balkema, Rotterdam.

Fidler, S., 2000, Bank schemes cost 2.6 m their homes, Financial Times, 26 September.

Houlder, V., 2000, Quarter of mammal species at risk, Financial Times, 26 September.

Husserl E., 1948, Erfahrung und Urteil: Untersuchung zur Genealogie der Logik, Claassen and Goverts, Hamburg //1973, Experience and Judgement, Investigations in a Genealogy of Logic, trans. Churchill, J.S. & Ameriks, K., Routledge, London. 

Jonoski, A., 2000, AquaVoice: A prototype for an internet distributed decision support system, 4th International Conference Hydroinformatics 2000, Cedar Rapids, IA, USA. (Proceedings available on CD ROM only: copies available from the author.)

Jonoski, A. and Abbott, M.B., 1998, Network distributed decision support systems as multi-agent constructs, Hydroinformatics ’98, Babovic & Larsen (eds), Balkema, Rotterdam.

Reisner M., 1993, Cadillac Desert, Penguin, USA

Roy, A., 1999, The Cost of Living, Flamingo, HarperCollins Publishers, London.

Sachs, J., 2000, The charade of debt sustainability, Financial times, 16 September.

Shiva.,V., 2000, Poverty and Globalisation, BBC Radio 4, Reith Lectures 2000 (http://news.bbc.co.uk/hi/english/static/events/reith_2000/lecture5.stm)

Thein, K.N.N. and Abbott, M.B., 1998, Internet-based management of water resources for a new Burma, Hydroinformatics ’98, Babovic & Larsen (eds), Balkema, Rotterdam.

Thein, K.N.N. and Abbott, M.B., 2000, The gender issue in hydroinformatics: some pragmatic aspects, 4th International Conference Hydroinformatics 2000, Cedar Rapids, IA, USA. (Proceedings available on CD ROM only: copies available from the authors.)

Watkins, K., 2000, Tweaking the figures will not help the poor, Financial Times,30 September / 1 October.

Yan, H., Solomatine D.P., Velickov, S. and Abbott, M.B., 1999, Distributed environmental impact assessment using internet, Journal of hydroinformatics, 01.1, pp.59-70.  3) The democratisation of decision making processes in the water sector

© "Journal of Hydroinformatics" www.iwapublishing.com

Part 2

M.B. Abbott
International Institute for Infrastructural, Hydraulic, and Environmental Engineering, Delft, The Netherlands,
and
Knowledge Engineering, BVBA, Brussels, Belgium

A. Jonoski
International Institute for Infrastructural, Hydraulic, and Environmental Engineering, Delft, The Netherlands

ABSTRACT : The second part of this paper investigates a possibility of realising an inversion of the so-far established order of knowledge/power structures in that it corresponds to an inversion in power relations that is realised by an inversion in knowledge relations. The system proposed here by way of an example is then primarily a means of realising this inversion. The economic sustainability of such a system within a ‘third world’ context necessitates the consideration also of knowledge/value relations, and these are also briefly introduced. The system itself is essentially a knowledge management system, comprising three principal components:

The use of ‘scientific discourse’ at the centre and of ‘narrative discourse’ at the outer periphery sets the overall specification of the inner periphery.

KEYWORDS : knowledge-intensive agriculture, advice-serving system, fact engines, judgement engines

1. KNOWLEDGE/POWER AND KNOWLEDGE/VALUE :

RELATIONS AND STRUCTURES

Knowledge-power relations are inherent in all human and many other animal societies, such as between parents and children of a human family and a and males and females in a wolf pack. These relations, however, become structured through habit and usage, finally often being codified in human societies in legal and other socially binding documents. In more recent times these relations and structures have been extensively researched in all manner of contexts, as exemplified in the well known works of Foucault (e.g., 1963//1973/1989 and 1966//1970/1989). Foucault saw knowledge and power as so inextricably entwined that he denoted them together as ‘knowledge/power'. One of the defining concepts of knowledge/power is the deliberate restriction and canalisation of knowledge so as to gain and maintain power and, conversely, the use of power so as to control flows of knowledge.

On the other hand, hydroinformatics, as a postmodern technology and thereby necessarily a sociotechnology, is tied into a postmodern condition where ‘knowers' transform increasingly into ‘consumers of knowledge' and knowledge, as a result, becomes increasingly a kind of product. It is then, however, not a product like just any other product, but one that enters into all other products. Thus increasingly many industries, agricultures and services, as well as many professions, become ‘knowledge-based' (Abbott, 1999a and b). This condition is then supported by such developments as those currently occurring in the use of the internet and in mobile telephony, so as to provide a ‘New Economy' that makes up a large part of the value of all companies quoted on most stock exchanges. Within this business context, however, knowledge exhibits another aspect that supplements and even surpasses its traditional, long-established, relation within knowledge/power, which derives from the following simple observation: that the same knowledge can be sold or exchanged indefinitely, without in any way depleting the original store of knowledge. Knowledge in this sense is an undepletable resource.

Now the knowledge of one individual or social group has always had, when relevant, a value to other individuals or social groups, so that it could be exchanged for other goods or for money. Thus knowledge has a value, and in a postmodern society this becomes in the very first place a value to a consumer of knowledge for whom it is relevant. To some extent, at least, this is a social value, for which this consumer pays in one way or the other in money terms. This is to say that a transfer of knowledge then comes to constitute a transaction, which itself entails a cost that is called a transaction cost. In a world increasingly populated by knowledge consumers who access, use and then for the most part discard knowledge, many bodies of knowledge can be sold or otherwise exchanged much more frequently than previously, so that the money cost of each application of the knowledge can be reduced. This decrease in the cost of each application of a given body of knowledge, even when augmented by new kinds of transaction costs, in turn increases the circulation of knowledge/value. It is by these means that knowledge acquires another, and in the longer run much greater, value as a society passes into a postmodern condition. In such a society, therefore, the social value of knowledge to its provider no longer depends only, or even so very much, upon maintaining knowledge differences or ‘gradients’ between the more empowered and the less empowered. It depends much more on the rate at which knowledge can be sold or otherwise exchanged and so made available, and then ultimately always as a means of improving the perceived quality of life of those who, as a result, buy this knowledge. Thus, in a postmodern condition of society, the determining factor in estimating the financial viability of any knowledge provider is the rate of circulation of knowledge/value. Every knowledge-intensive activity, such as a knowledge-intensive agriculture, must base its financial viability on this measure. Thus, in so far as a large, and even very large, number of end-users can be brought to make use of the same repository of knowledge, the application costs for each end-user can be brought down to levels that are compatible with the advantages conferred by this knowledge. 

At the same time, however, values are by no means necessarily measured in social value terms: intrinsic values, usually associated with the emotions that the individual experiences in giving or receiving knowledge, are often more important to overall institutional sustainability than are social, including money, values (Abbott, 2000b). Thus, for example, even in current ‘corporate business’ applications, incentives to share knowledge with others in order to increase social value are commonly introduced by setting up gradients in intrinsic values. Closer physical proximity, enhanced means to promote sociability, and the encouragement of peer respect for those who contribute knowledge, are already well-known devices for promoting knowledge sharing in established business environments. For our part, we observe that ‘creative business' begins when knowledge/value relations transform into knowledge/value structures in which value necessarily takes on both social and intrinsic attributes (Abbott, 1999a and 2000a). In a case like the present one, that is directed to empowerment by knowledge provision, it may be useful to think of a triadic construction of ‘knowledge/power/value' and the translation of this from a relation into a structure through a sociotechnical process.

2. KNOWLEDGE-INTENSIVE AGRICULTURE

A knowledge-intensive agriculture is then one that can only be sustained through a high rate of consumption of knowledge. It is correspondingly an agriculture that can use ‘the same knowledge' so very often that it can both sustain the provider of this knowledge (who remains a ‘knower’, but now in an exponentiated sense) and bring down the cost of each application of this knowledge to a level that is justified by the gain that is thereby conferred upon its consumer in income and, inseparably from income, in empowerment. The advantage to the farmer of this arrangement is that this individual, family or other social group can then farm in ways that would not be feasible without this supply of knowledge. Now it can be no ‘historical accident' that nearly all the forms of sustainable, ‘ecologically responsible' or ‘organic' farming that so proliferate around the world at the present time are marked by a need for ever greater rates of application of knowledge and associated understanding. For example, contributors to the Indus Valley internet forum speak of employing some 12,000 species in their sustainable farming practices, with one farm typically using some 100-150 species. This is necessary in order for farmers to obtain a balanced diet for themselves and for their animals, to provide habitat and nourishment for other creatures, such as bees and predators on plant-destructive insects, to provide adequate cash crops and to maintain the nutritional balance of the soil. Sound irrigation and drainage practices and other soil-cultural practices that depend upon water to promote a justifiable productivity further necessitate knowledge relating these aspects to those associated with plant cultivation. It is at this place that hydroinformatics has such a vital role to play. Although types of crops and their yields vary widely over the great number and varieties of farming experiments and experiences of this type, estimates of increases of between two and five times in yields per soil area are regularly reported, together with much improved balances of diet, but then with little or no use of industrial-agrochemical inputs. Levels of mechanisation also tend to be considerably lower. On the other hand, most such agricultural practices are considerably more labour-intensive than those prevailing within the current Western paradigm, necessitating that a substantially larger part of the total working population is employed in agriculture. Although this part is so much larger than that of the most ‘advanced' agricultures in Europe, not to speak of North America, this may be seen as an advantage in terms of the alleviation of poverty in many ‘third-world' societies. Incomes in dollar terms can be expected to be considerably lower than those of farmers surviving the current round of industrialised agricultural ‘consolidation' in North America, with the result that the advantage in increasing purposeful and sustainable employment can only be sustained economically when this labour force is properly informed and advised.

3. A POSSIBLE SOCIOTECHNICAL REALISATION OF AN ADVICE-SERVING SYSTEM

As mentioned earlier, the system as it is envisaged here is a sociotechnical system, and for its successful realisation it is necessary that both aspects, the social and the technical, are woven together in a carefully studied and analysed way. The crucial goals are to be achieved only if there is a recognition of the fact that the social and the technical are so interconnected and mutually dependent that they have to be approached and dealt with in a fully integrated fashion. Thus it in no way suffices to study only the social consequences of technological innovations or the technical innovations required to realise certain social changes. Sociotechnical studies of this kind do not belong to the category of social studies anymore than they belong to the category of technical studies: sociotechnical developments are qualitatively different from the social and technical developments from which they are composed. To follow an ancient analogy, the social is as one piece of thread and the technical is as another piece of thread, but the sociotechnical is as the piece of cloth that is produced when these two pieces of thread are woven together. The cloth is qualitatively different and has an entirely other use and value than the threads from which it is woven. Following this understanding, in the subsequent sections both the social and the technical aspects of the three system components will be treated simultaneously, even though they must be introduced in the order of writing as though they were separate.

Looking at the technical side, it is necessary first to introduce a general technical approach that may be applied for the conceptual design of the whole system. At its implementation level, such a system may advantageously employ many different and relatively advanced technologies of kinds that have been extensively researched and applied in recent years. However, regardless of the actual technical implementation, on a conceptual level the proposed system is most appropriately described using a framework based on agent orientation using message-passing architectures. The overall system design may then also be based on this conceptual framework. There are of course many definitions of ‘agent’ and ‘agenthood’ in the technical literature, but for our present purpose we may regard an agent as any software entity that combines methods and data such that it can initiate and perform certain tasks within a larger, system, context. Although agent-orientated architectures are usually associated with loosely coupled distributed systems with emergent properties, they will be introduced within the present context in a more limited and restricted way.

With respect to the tasks to be performed by the agents of this system, we may propose three basically different kinds of agents, with each kind associated primarily with one of the system components. The first one takes up and carries the physical, social, intentional and other features of the individual user of the system, for the most part situated within the outer periphery, and communicates these in the form of intentions and narrative knowledge, to the second kind of agents, situated for the most part within the inner periphery. This second class of agents then makes use of the information so communicated to pass on narrative knowledge translated into a scientific form suited to the workings of the centre. This is then used to synthesise judgements upon advisable courses of action to be taken by the individual user subject to all the knowledge and data that the knowledge centre can provide. This advice is then carried back by the first class of agents to the individual user in forms that this user can readily assimilate. The third class of agents is located in the knowledge centre and is responsible for maintaining the necessarily high level of software interoperability of the various software components (systems for representing of different classes and forms of narrative knowledge, different kinds of models, databases, software modules for extracting measured data, modules for satellite image interpretation, etc.). On requests from the inner periphery, these agents have to provide the requested facts by invoking whatever software component is necessary and then deliver the data back to the second class of agents in the inner periphery in appropriate form. In current ‘business speak’, the agents of this third kind are primarily concerned with ‘supply chain management’, albeit of knowledge rather than of material things. In this case, much of the same vocabulary – ‘supply networks’, ‘parallel chains’, ‘enhanced concurrent activities’, ‘synchronised supply models’, etc - can be taken over from established business practice, even though applied in quite different ways..

In this respect, and as observed earlier, the systems envisioned here, although in essence knowledge management systems, proceed in almost the opposite direction to those taken by the knowledge management systems that are at the centre of most current interest. The current dominant paradigm in knowledge management is one in which the collective knowledge of an organisation, often called its ‘corporate knowledge’, is made accessible across the organisation in a manner and to an extent that is defined and controlled by corporate management. This practice can be seen as a natural extension of corporate information management and it makes use of many of the same tools by way of enabling technologies. In systems of the type considered here, on the other hand, there should be as little ‘corporate’ management as possible of this kind. The current paradigm has to do with such issues as ‘Enterprise Resource Planning’ (ERP) and ‘Enterprise Asset Management’ (EAM), while the paradigm introduced here is directed to providing knowledge, initially in the form of advice, to large numbers of individuals, families and other social entities. We might indeed try to emphasise this difference by describing the systems that are proposed here as particular kinds of ‘knowledge self-management systems’. The orientation towards agent-based architectures in this class of systems can then be seen as a technological response to their quite different social intentions.

(We might proceed even further to describe these constructs as particular kinds of ‘self-organising sociotechnical systems’. This however has the consequence that, since they are intrinsically nonlinear, they must be expected to exhibit chaotic behaviour in the strict scientific sense, which is to say that they must be creative of an underlying order (Lorenz, 1993; Ruelle, 1993; see also, within the management context, Anderla et al, 1997). Thus the macroscopic analysis of these systems may be expected to require an extension of existing chaos theory (see, further, Kaufmann, 1993)). 

The roles of these three classes of agents will be further discussed in the following subsections, weaving them into the expected social context. It may be anticipated that most of the challenges to appear from the social side will be in the outer periphery of the system, whereas the development of the inner periphery and the knowledge centre (with its relations to the external knowledge providers) will be facing predominantly technical problems.

3.1 The outer periphery

This is by far the most difficult part of the system to define at the present time. The main reasons for this are the following :

  1. Most processes at this periphery proceed currently through experience, intuition, custom and tradition, commonly held in place by religious rituals. These interacting knowledge structures, many of them based upon so-called ‘tacit' knowledge, are also endangered by such recent developments as the introduction of technocratic irrigation and drainage practices and genetically modified crops.

  2.  
  3. Being for a large part intuitive, these knowledge processes occurring at the periphery of the system are rarely verbalised by the persons directly concerned; indeed in many cases (and as commonly occurs in the case of using tools) verbalisation cannot suffice of itself, but must be accompanied by repeated exemplification. This and other constraints necessitate that the systems proposed for introducing the new paradigm must be supported by knowledgeable persons at the field level, so as to make them ever more definitively sociotechnical systems. The basic reasons underlying the preponderant role of women in staffing these systems, as has already been experienced by microbanking and telecommunications providers, has been investigated in Abbott (2000a) and Thein and Abbott (2000).
  4. Systems of the type that have so far been proposed for sustainable ‘organic' farming have necessitated levels of co-operation between their users that are in many ways beyond those of existing traditional agricultures, and even these are being further undermined by the ongoing intrusion of the technocratic paradigm. Thus the proposed new ‘knowledge-intensive' systems must be designed also to promote changes in attitudes in the direction of more collaborative attitudes and enhanced social cohesion.

  5.  
  6. Similarly, following Heidegger's celebrated analyses of tool-using processes in general (see Abbott, 1991) there can be no ‘theory’ of these processes but, at most, a theory of their possible descriptions and corresponding reductions, as ‘phenomena’. Correspondingly, we regard the Husserlian/Heideggerian school of phenomenology as the underlying science of all such, sociotechnical, processes. Thus, even the existing relations between intentional acts and intentional objectives can only be made explicit generally within this kind of phenomenology.

  7.  
  8. `Giving advice’, implying as it does the changing of intentional acts, and occurring together in many cases with the introduction of new intentional objectives, necessitates a careful analysis of the existing phenomena and the changes that the adoption of this advice may produce. This analysis must then proceed, initially at least, through ‘the phenomenological method’ of extracting the essential features of processes from careful and profound studies of a wide range of practical situations, as ‘examples’ or ‘case studies'. It should be emphasised again here that we are using such expressions for what they correctly denote, but not with the connotations most commonly associated with them at the present time. Thus, these are not ‘case studies’ in the usual sense of schools of business administration, for whom ‘case studies’ are means to understand the dynamics of social-organisational situations, or ‘group dynamics’ involving only persons. Here we are again dealing with something that is qualitatively different, which we might try to describe as the ‘sociotechnical dynamics’ of intersecting and interacting social and technical processes, as the warp and woof in the sociotechnical weaving process. This proceeds for the most part between people and equipment and, through the intercessions of both of these again, with the physical world. 

  9.  
  10. Thus, although we may speak loosely of ‘narrative discourse’, and even ‘narrative knowledge’, the process of narration itself may take oral, textual, graphical, repetitive-exemplary and other forms. Accordingly, this ‘discourse’ may be extended and often quite diffuse, and the ‘meaning content’ of the narrative knowledge obscure and ill defined by conventional predicate-logical, modern-scientific, standards. This implies that the applications of the established phenomenological methods will have to proceed iteratively, building further at every stage on their earlier results. In terms of classical Meinongian object/value theory, all such objectives will have to be ‘implected' (Abbott, 2000a).

  11.  
  12. These many basic differences between what is being attempted here and existing practice in the dominant corporate business environment imply that most of the existing literature on business studies, and on ‘knowledge management’ in particular, has little to say that is relevant to the present purpose. Certainly, the existing literature can have no use for studies in phenomenology, hermeneutics and existential philosophy. Indeed, such studies would be subversive of the whole thrust of the existing corporate paradigm, exposing its essential emptiness in human terms. Thus, as Heidegger first pointed out so long ago now (1927//1962) and subsequently amplified, technology in the modern era is no longer the instrument of mankind, but it is mankind that has become the instrument of a certain distortion of technology, popularly called ‘modern technology’, so that mankind itself has become something ‘technocratic’ (see also Abbott, 1999b and 2000b).
Although this situation is complicated further on its physical side by the sheer number and variety of the cases that have to be considered, it can also be much simplified in practice by drawing upon the facilities that have already been introduced by the existing micro-banking, mobile telephone and related networks at the village level. By the same token, the very success of these institutional networks in providing money-credit and communication facilities at the village scale indicates that the problems of introducing advice-serving arrangements at this scale are also solvable. Indeed, it is anticipated that many of the same persons and arrangements may be carried over from the credit-providing and repayment and telecommunications-providing roles to the advice-providing, advice-discussion and ultimately, advice-payment roles.

On the basis of this foundation in people, methods and technologies, and following existing and well-proven phenomenological methods, it is proposed to construct both the judgement engines (located in the inner periphery, and to be described shortly) and the means employed for their instantiation. This will proceed on the basis of phenomenological reductions of considerable numbers of field examples, regarded as ‘case studies' in the sense described above. It is anticipated that this procedure will also play a valuable role in identifying the most relevant examples.

Microbanking and other such organisations have immediate interests in applying such advice-serving systems within the context of their own existing agricultural development foundations, since such services will immediately provide means for better protecting their loans and other investments. This means that in several knowledge/value-sensitive areas they need only move over to money payments for such advice as the self-confidence of farmers in their ability to use this advice and their trust in its relevance and reliability become better established. This can proceed initially by way of prototypes in those sociotechnical environments where the greatest return from the availability of sound advice is to be anticipated. At the present initial stage, however, a more general analysis of the phenomenology of these sociotechnical processes has to be conducted and for the moment this is proceeding through academic studies carried out by nationals of the countries in which applications are most likely to succeed. The intention, however, is to proceed to field applications as the enabling technologies become increasingly, and increasingly cheaply, available.

Turning back now to the technical approach for achieving the tasks in the outer periphery, the first thing to be considered is the existing infrastructure of the telecommunications network which already exists at the village level. At this level there are still only a very few knowledge processors (often still called ‘computers’) which are cable connected and may be used for direct land-line communication with the inner periphery and further with the knowledge centre. This directs the design of the first class of agents located in the outer periphery initially to very ‘light’, purely message passing agents suited to the limited capabilities of current Wireless Application Protocol (WAP) technologies and so with only two primary tasks or ‘responsibilities’. In the first direction of knowledge transmission these agents have to take the features and knowledge inputs of each individual user (physical, personal, social and others) as a message to the inner periphery. Upon completion of the requests through the actions of the agents in the knowledge centre and the inner periphery, these agents have to bring the advice, translated into narrative form, back to the individual user. We shall return later to the way in which this narrative form may be realised (voice message, written message, visual cartoon, etc.). Many of the features associated with this function appear to be only fully realisable with the introduction of complete ‘third-generation' (‘3G') technologies, as anticipated by the years 2002-2003, although the provision of simple base stations supporting a local wireless broadband computer network may suffice to support WAP enabled technologies in the meantime.

Since the first kind of agent here has the function only of carrying the truth-bearing messages of others and has, so to say, ‘no voice of its own’, we currently call such agents ‘angels’. In conventional studies of agent-orientation they are often called ‘flag agents’, being said to ‘flag-in’ when activated and ‘flag-out’ when they deactivated themselves. Although we personally prefer the name of ‘angels’ (and even ‘guardian angels') within the present context, because of the emphasis that we wish to place upon the truthfulness and personal relevance of the messages that these agents carry, we have to accept that many persons, not being acquainted with theological studies of angelology, may find this usage objectionable (see, for example, Barth, 1944-1953//1960, pp 369-418). Correspondingly, we shall refer to these agents simply as messenger-agents. The message that they deliver, insofar as it contains the description of the needs for knowledge of the end-user and the ability of this user to employ this knowledge, is called the user profile of that end-user. (See Abbott, 2000a).

3.2 The inner periphery

The second class of agents is composed of altogether ‘heavier’ and more earthly entities. If we may now turn to use the kinds of biological metaphors that are so widely employed in software engineering and artificial intelligence, we might say that these ‘heavier’ agents are ‘impregnated’, ‘fertilised’ or ‘seeded’ by the lighter and purely message passing, agents. They are agents that search the facts when they are so impregnated with the messages carried by the messenger-agents so as to collect just those facts that are the most relevant to the individual user. They then interpret these facts in such a way as to arrive at advisable courses of action for the individual users, as described later. Although we personally refer to these heavier, more earth-bound, agents as ‘saints’, we may less controversially describe them as ‘relational agents'. In more conventional software engineering terms, the impregnation or seeding first defines a navigation path through the facts and then defines the interests to which these facts are to be directed relative to the intentions and capabilities of the individual user, as also communicated through the impregnation by the user’s individual messenger-agent with that user's profile.

We observe that although the number of messenger-agents may run from thousands into millions, their individual visitations may be quite few and far between. On the other hand, the number of relational agents that are subject to these visitations may be relatively small, even though each of these relational agents may have much more work to do, thereby taking a lot more time over responding to each visitation. The timing of visitations is clearly of great importance as it must be co-ordinated with periods of critical decision making by the individual user for whom the messenger-agent is, so to say, a ‘guardian'.

The ‘ facts of the matter’ must then be considered upon each visitation in relation to the profile of the end-user of the system. For this purpose the relational agents may be expected to make use of ‘judgement engines’. These are devices that facilitate the making of judgements in general and in this case of judgements about the best advice that can be given to an end-user in the fact-given situation (Abbott, 1999a and c; Yan et al, 1999). A judgement engine is a device that realises an inference string that has as its simplest form: 

(beliefs, facts) ® attitudes ® positions ® judgements ® decision ® action (1)

in which the beliefs are provided in the form of the user profile by the messenger-agents, while merging of these beliefs with the facts is the task of the relational agents. The relational agent sets up or ‘instantiates’ a judgement engine on the basis of the end-user profile. This is to say that the relational agent introduces the kinds of facts that are considered important by the end-user and the beliefs of that end-user concerning the desirability or otherwise of a range of possible facts within that user’s personal situation. The knowledge provided by the end-user is also processed through the relational agent into a form whereby it can be assimilated by the centre.

Thus, to take a simple case, the value that a rice farmer might place upon each particular depth of water over his fields, at a given stage in the rice-culturing process, can be used to express the farmer’s beliefs concerning the value that he (or she) would put upon a certain water depth were it to become a fact. In effect, in this way the judgement engine elaborates a belief-value-to-fact mapping, or function, of a kind that is conventionally represented by the curve of a graph. The water depth as actually predicted by the knowledge centre can then enter as a fact and this transforms through the ‘fact-value’ mapping or function into a value to the end-user. This is of course only one value among many others and the way in which the different results are aggregated according to the various criteria is a favourite subject of ‘multi-criteria analysis’ in economic theory. There are of course correspondingly many ways of conducting this analysis, of which three are employed in current engines. (See, for example, Yan et al. 1999, Bazartseren et al, 1999 and Jonoski 2000). Among other features, one of these methods allows for a calculation of the inconsistencies that are commonly inherent in the belief system of the end-user.

We may consider the relational agents as much extended ‘search engines’ which now, however, seek out facts and interpret these facts in the light of the profile of the end-user so as to provide this user with the best advice that they possibly can. The relational agents must then have access to the facts. All the devices that are used to provide the physical, biological and other such non-personal facts of the matter using all available narrative and scientific means are called fact engines. Being situated within the inner periphery of the system, the relational agents can draw directly upon all the productions of the fact engines of the knowledge centre for this purpose. Moreover, since the inner periphery will mostly be cable-connected to the knowledge centre, it can draw upon the outputs of all fact engines and otherwise make use of the facilities of the centre through the system intranet. This intranet can also be used by the relational agents to transmit the narrative knowledge that they have garnered at the outer periphery to the centre in a scientific form. The intranet must then comply to the same standards as the World Wide Web (WWW). The facts may thus be assembled at the centre where they can be merged with the narrative knowledge garnered at the outer periphery and transmitted to the inner periphery using, for example, standard Geographical Information Systems (GIS) platforms.

3.3 The knowledge centre

The knowledge centre constitutes the knowledge core of the system. It is the most ‘earth-bound' element of the system, being reduced to the ‘mundane' application of scientific methodologies. It responds to the requests of the inner periphery for information, which is then so far as possible provided by the Web GIS systems. It must itself then prepare the material for transmission, on the one side by running models, interrogating measurement stations, requiring and interpreting satellite images and through other such scientific and technological activities, and, on the other side, by merging all this with narrative knowledge expressed in a suitable scientific format. Since however, by far the greater part of the scientific material can already be supplied by existing organisations, these become in their turn ‘knowledge providers’ to the centre. It is these that we have called the ‘external providers’ earlier here. It can be expected that the required high level of efficiency and interoperability of the various fact engines, located either in the knowledge centre or at some of the knowledge providers, will best be achieved if these fact engines are reconfigured so as to better comply with the requirements of the agent-orientated approach. From the perspective of this system they would then all fall into the third class of agents which we shall here call ‘service’ agents. The system intranet then becomes a platform on which such service agents operate in order to provide the requested knowledge and (facts(data)) from the inner periphery. Once again the WWW becomes the instrument for enabling the provision of such knowledge and data, or more generally ‘information’, or more generally again ‘content’. When seen in this perspective, the knowledge centre takes on the attributes of a ‘knowledge broker’ whose principle task is one of knowledge management in a rather more conventional sense. This role is strengthened further to the extent that a return flow of information is engendered from the outer periphery and transmitted through the inner periphery to the centre. In this case the knowledge centre can provide knowledge back in turn to what are otherwise its external knowledge providers. The door is thereby opened to a far superior relation between the work of the ‘providing’ organisations and the real world situation, or ‘the situation on the ground’, and this provides in turn a considerable added value.

The envisaged functioning of the system, achieved by the bi-directional flow of knowledge through the three system components introduced in the previous subsections, is schematised in Figure 1.


Figure 1. Schematisation of bi-directional knowledge flow in the proposed system

4. TECHNOLOGIES FOR THE DEVELOPMENT OF THE SYSTEM

Even with the introduction of the system so far, it is clear, as observed earlier, that the design and implementation of it will require the application of some rather advanced Information and Communication Technologies (ICTs). The tasks of the three different classes of agents operating within the three respective system components vary considerably, and it is obvious that a combination of different technologies will be required for their design and implementation. 

As the connections between the outer periphery and the inner periphery will be for the most part wireless, the messenger-agents will have to make use of the emerging wireless internet technologies. At present, the Wireless Application Protocol (WAP) has appeared as a standard upon which wireless internet communication is mainly based. This protocol enables communication between the client (a wireless device such as a mobile telephone) and a WAP server connected to any number of other Web servers. The WAP server is known as WAP gateway, which routes clients’ requests to HTTP (or Web) servers. The communication is based on the so-called Wireless Mark-up Language (WML) which adheres to the XML (eXtendable – or ‘eXtensible’ - Markup Language) standard. Although there are ways of translating HTML-coded information into WML, because of the severe limitations of the present wireless devices (low bandwidth, low memory, limited screen size, etc.) most services are now being developed separately, either as static WML documents, or dynamic WML documents created on the server side. The dynamic WML documents can be generated with any server-side technology (CGI scripts, Active Server Pages, Java Servlets, Java Server Pages, Perl, Tcl, etc.). These documents are then sent to the client again through the WAP Gateway. 

In the proposed system the WAP gateway will be located in the inner periphery, and it will be the task of the relational agents to handle the requests coming from the outer periphery through this gateway. The handling of the requests is, however, much more complex as compared to the provision of a simple service through a WAP server. In order to provide the relevant advice as a response to the outer periphery, the relational agents will have to initiate another communication process between the inner periphery and the knowledge centre. Upon receiving the relevant facts from the knowledge centre the relational agents will then be responsible for instantiating the judgement engine applications that are set up with the profiles of the individual users in order to assemble a ‘best advice’ for each of the individual users. After translation into narrative forms, this advice will be sent to the outer periphery through the WAP gateway. The translation may proceed through existing standard procedures for translating message strings (commonly in the form of binary strings associated with strings of natural numbers) into linguistic labels and assembling them into texts. The texts may be transmitted in textual form or synthesised into a vocal form, running together with illustrative material if required. Again, the form of delivery will depend crucially on the possibilities offered by the wireless technology, and with the arrival of the third generation (3G) wireless devices many of the present limitations may be overcome.

Even at this most basic stage there is already a vast panoply of tools for realising and facilitating the construction of such systems. In the direction of Computer Aided Telephony (CIT) these tools normally observe the need to satisfy the so-called asynchronous transfer mode (ATM) standard, providing major advantages at the moment only for cable transmission. Client profiling and advice serving can already benefit from the introduction of subscriber identification module (Sim) cards – the tiny smart cards in handsets which identify the customers and control their relationships with the network – as exemplified by the Virgin Mobile service. Of considerable potential significance here is also Ericsson’s Bluetooth technology, which provides short wave radio connections between a wide range of devices and a computer network. It is to be expected that most advice serving systems will use Advanced Speech Recognition (ASR) and Text-to-Speech (TIS) software. Although the voice-enabled services thus provided can be accessed through web browsers, such as is in turn enabled through the use of XML, more powerful tools, such as Motorola’s Vox ML can be used. Typically for such tools, VoxML further provides a software development kit and a web site to promote it. Developments along these lines will be greatly strengthened through an industry-standard voice browser technology, such as Voice XML, which is currently being supported by several of the largest ITC companies. The new ‘talking web sites’ require new types of web servers, as represented by Lucent Technologies’ SpeechServer, which connects directly to the service provider’s database and to the (voice accommodated) internet.

The complexity of this whole process will not allow a design and implementation of the relational agents as simple ‘middle tier’ applications located between a client and a server. In fact the whole inner periphery cannot be reduced to completely automated unmanned part of the system. Instead, it is proposed here that the inner periphery with the knowledge centre are together organised as the system intranet. This system intranet will have two extranet parts on the two respective ends. The communication through the WAP gateway with the outer periphery may be seen as the first extranet part, while at the other end the communication between the knowledge centre and the various knowledge providers may be regarded as the second extranet part.

The inner periphery and the knowledge centre, forming the system intranet, will be cable connected and the most obvious approach to their design is to use the existing technologies developed for the WWW. Standard HTML, scripting languages like JavaScript, various Java technologies, CGI, Active X, and other technologies are already utilised by organisations within the countries towards which these systems are directed, and already for a variety of purposes: the experience and knowledge is already present locally. Once again, there is a great variety of Customer Relationship Management (CRM) tools available on the market that might conceivably be adapted to the present purposes. The same applies to ERP and EAM applications. It is not at all clear at present, however, whether or to what extent these tools can be so adapted advantageously.. 

One of the more important aspects for the successful exchange of data and knowledge between the inner periphery and the knowledge centre will be the WWW-GIS platform which will enable quick and reliable retrieval of the requested data at the inner periphery as these are delivered by the knowledge centre. The retrieved data can then enter as inputs to the judgement engine applications and combined with the user profile and request in order to provide the necessary advice. At the same time, the retrieved data can be easily visualised at the inner periphery through a browser environment. 

For this purpose, a full utilisation of the recent technologies for deploying GIS on WWW, such as the Open Geospatial Datastore Interface (OGDI) which uses a Geographic Library Transfer Protocol (gltp) seems appropriate. This interface provides standardised methods through which software packages can access, ‘on the fly’, a variety of geospatial data products. OGDI uses a client/server architecture to facilitate the dissemination of geospatial data products both locally and over any TCP/IP network, and a driver-orientated approach to provide access to many geospatial data products and formats. There are many advantages to OGDI, as set out on (http://www.globalgeo.com): 

The role of the WWW-GIS platform is stressed here because most of the data will be geospatial and this approach is then most suitable on the side of the working of the inner periphery together with the centre. At the centre, for its part, the Web Server will also be connected to the various fact engines in the knowledge centre itself and at the separate external knowledge providers. These external sources will also provide non-geospatial data and will translate the requested facts in Web-compatible format so that they can be received at the inner periphery. Some of these other data will as well be necessary inputs for the judgement engine applications, but they can in addition carry descriptive information, which may be directly transferred to the outer periphery so as to support the recommended advice. 

At the knowledge centre itself, the WWW-GIS platform which will enable the communication between the knowledge centre and the inner periphery will depend on the fact-gathering tasks which are carried out by the service agents. The main problems related to these tasks are those of efficient interoperability of different fact engines. Some of the fact engines will be available at the system intranet, and relatively easy integrated with the WWW-GIS platform. However, as mentioned earlier, the role of the knowledge centre will be more determined by its tasks of ‘knowledge brokerage’, when fact engines located outside the system intranet in various existing external organisations are employed. The issue of software interoperability then becomes more complex, as it relates to integrating legacy software, which is commonly software developed by different people in different languages and for different platforms. 

The software interoperability problem has been already addressed by a number of existing technologies. The integration of the client-server model, which underlies the communication and computation over electronic networks, with the object oriented technologies, has resulted in a variety of developments, all of which may be gathered under the topic of ‘distributed objects’. The best known of these are Microsoft’s DCOM (Distributed Component Object Model), CORBA (Common Object Request Broker Architecture), developed by the consortium OMG (Object Management Group), and Java’s RMI (Remote Method Invocation) (see Yan et al, 1999). All of these technologies have their advantages and disadvantages, but the basic idea behind all these approaches is to package software into components (objects), with well-defined interfaces through which they can inter-operate and invoke each others’ methods. This interoperability can be achieved in different client-server modes, and, with some technologies, even across different operating systems. In order to extend this software interoperability to the existing legacy applications, these will have to be provided with the appropriate interfaces. 

Any of the above mentioned ‘distributed objects’ technologies may be used for the development of a framework for integrated operation of the service agents with the various fact engines. However, the task of equipping the ‘legacy’ fact engines with the appropriate interfaces for operation in such a framework remains extremely difficult, because it requires a lot of low-level interventions in these applications. This task is complicated even more when these fact engines are coming from different knowledge providers. An alternative approach to this problem is to adopt an implementation framework which will closely follow the Agent Orientation Programming (AOP) approach (Shoham, 1997) and link the service agents and the fact engines with a highly expressive, specially developed Agent Communication Language (ACL). The ACL which has already emerged as a standard for this purpose is the Knowledge Query and Manipulation Language (KQLM), developed at Stanford University, USA (Finin et al., 1997). This approach although advocated by many developers of so-called Multi-Agent Systems (MAS), has still many unresolved problems (see especially Nwana and Ndumu, 1999). The most limiting one is certainly the so called ‘ontology’ problem which is related to the need to have specifications of conceptualisations (ontologies in MAS terminology) of the problem domain which are common to, and shared by all the agents in the framework. Another alternative which has appeared recently, is to provide the service agents with a linking framework by, again, using XML technologies. XML is designed as a metalanguage for describing data and documents, thereby providing a framework for easier interoperability of various applications. Future research will be very much orientated towards investigating these alternatives.

As for the choice of a particular technology for the technical implementation of the agents associated with all three components, it may be expected that many possibilities will be considered and the final choices will have to be made during the phase of the system design. The most recommended choice from the beginning however, is to make use as much as possible of the various Java technologies. Java was designed from the outset with network programming in mind, and nowadays is probably the best choice for programming distributed applications.

CONCLUSIONS

The rapidly advancing storm of public opinion that threatens water professionals is in many cases justified by events, and the water professionals themselves cannot avoid much of the blame that now increasingly rains down upon them. In this situation these professionals must seek ways to better satisfy the needs of the societies which they serve. To the extent that these societies now pass increasingly into postmodern conditions, where knowledge becomes a central value and the population as a whole becomes ‘consumers of knowledge', this response of the water professionals must be in the direction of providing knowledge to the general population. This knowledge provision must then serve the interests of the population as a whole, empowering the individuals, families and other social groups within this population as genuine stakeholders in water resources, while also enhancing their awareness of their environmental responsibilities. This can lead to more equitable and sustainable distributions and uses of water. To the extent that so-called ‘third-world' societies move in this direction, such as through introducing knowledge intensive agricultures, knowledge-intensive aquacultures, knowledge-intensive health service provisions and other such services, so they will have to apply knowledge management systems of the type that is introduced here by way of a possible prototype. Clearly, the realisation of these systems will require new institutional structures, necessitating the disaggregation and re-aggregation of existing social, including agricultural and industrial structures, and this has also to be researched. The realisation of this new paradigm now appears as the greatest challenge facing hydroinformatics on its sociotechnical side.

The sustainability of systems of this kind clearly necessitates another level of coexistence, and even in some cases co-operation, between different end users. This necessitates that there is at least some understanding of the judgements and actions of others and, beyond this again, that there is at some stage a change of attitude towards ‘the others’ in a society. Both of these developments can be considered in relation to the dual of the category shown in Eq.1:
 


action ® decision ® judgement ® position ® attitude ® (beliefs (facts)) (2)

where now actions and facts are observables. In the first case this represents the process of understanding why a neighbour or some other such person of influence in the life-world of the individual end-user behaved in the way that he or she did in a given instance. In the second case it has the interpretation that the performing of actions increases the awareness of the consequences of an own decision-making and judgemental process, thus providing experience Connecting back experiences to the judgements that produced those experiences, the positions and thus attitudes may be brought into question and, under the influence of repeated experience, the attitudes themselves, and thence the belief system, may change (see, especially, Husserl, 1948// 1973). The first of these interpretations can be realised using logical discourse, while the second is usually supposed to proceed at the level of the pre-linguistic and pre-predicative, so that it is largely the product of a more extended, personal, experience of using such a system in the manner of a tool.

It is a basic tenet of hydroinformatics that every new development must not only provide ameliorations in the life of people and their natural environment, and these inseparably, but it must also provide business opportunities. In the present case these opportunities appear as accruing to what is often called the ‘New Economy’, of the information and communication technologies, and in many cases in opposition to the ‘Old Economy’, associated here with genetically modified crops, agro-chemicals, and to some extent also with mechanisation. The new paradigm that is advanced here may then be envisaged, at least on the surface of things, as a confrontation between the ‘Old’ and the ‘New‘ economies, where it is to be expected that the ‘Old’ economy will try to take on many of the clothes of the ‘New’, as already exemplified by the ‘new rhetoric’ of the World Bank and other such organisations. This then constitutes a further part of the ‘business challenge’ of this initiative. Although the business development of the present initiative is also essential to its practical realisation, it does not seem appropriate to introduce this aspect at this juncture.

As is usual in all sociotechnological studies, there is a synergy here between technical advances that appear to be very new, even as their roots run far back into the past, and the underlying current of ideas, which is often very old but which carries its message with undiminished, and indeed amplified, relevance into the present time. Here, as in so many other places, the past is not only irrevocable; it is also irreplaceable.

REFERENCES

Abbott, M.B., 1991. Hydroinformatics: Information Technology and the Aquatic Environment, Aldershot, UK/Brookfield, USA: Ashgate.

Abbott, M.B., 1999a, Introducing hydroinformatics, Journal of hydroinformatics, 01.1, pp. 3-19,.

Abbott, M.B., 1999b, Forchheimer and Schoklitsch: a postmodern retrospection. Proceedings of XXVIII IAHR Congress, Graz, Austria. 

Abbott, M. B., 1999c, Hydraulik im Zeitalter der Hydroinformatik, in Steinebach, G., ed., Mathematische Modelle in der Gewaesserkunde, Bundesanstalt fuer Gewaesserkunde, Koblenz. (In German).

Abbott, M.B., 2000a, The gender issue in hydroinformatics, or Orpheus in the Underworld, Journal of hydroinformatics, 02.2, pp. 87-104. 

Abbott, M.B., 2000b, Fides et Ratio: a critical appreciation. Journal of hydroinformatics, 02.2, http://www.iwaponline.com/jh/002/jh00200e1.htm.

Anderla, G., Dunning, A., and Forge, S., 1997, Chaotics, Academic, London.

Barth, K., 1944-1953, Kirkliche Dogmatik, Evangelische, Zollikon-Zurich// 1960,Church Dogmatics, transl. Bromiley, G.W., and T.F.Torrance, Clark, Edinburgh, Part III, Vol. 3..

Bazartseren B., Solomatine D.P., Minns, A.W. and Abbott, M.B., 1999: Improving the quality of judgement in environmental impact assessment by using an analytic hierarchy process, Proceedings of XXVIII IAHR Congress, Graz, Austria.

Finin, T., Labrou, Y., & Mayfield, J., 1997, KQML as an Agent Communication Language, in Software Agents, J.M. Bradshaw (ed.), MIT Press. 

Foucault, M., 1963, Naissance de la Clinique, Presses Universitaires, Paris//1973, The birth of the clinic, Tavistock, London/1989, Routledge, London.

Foucault, M., 1966, Les mots et les choses, Gallimard, Paris//1970, The order of things, Tavistock, London/1989, Routledge, London.

Heidegger M., 1927, Sein und Zeit, Neimeyer, Tuebingen.//1962, Being and Time, transl. Macquarrie J. and Robinson E., Blackwell, London.

Husserl E., 1948, Erfahrung und Urteil: Untersuchung zur Genealogie der Logik, Claassen and Goverts, Hamburg //1973, Experience and Judgement, Investigations in a Genealogy of Logic, trans. Churchill, J.S. & Ameriks, K., Routledge, London. 

Jonoski, A., 2000, AquaVoice: A prototype for an internet distributed decision support system, 4th International Conference Hydroinformatics 2000, Cedar Rapids, IA, USA. (Proceedings available on CD ROM only: copies available from the author.)

Kauffman, R., 1993, The Origins of Order, Oxford Univ., Oxford.

Lorenz, E., 1993, The Essence of Chaos, UCL, London.

Nwana, H. and Ndumu D.,1999, A Perspective on Software Agents Research, Knowledge Engineering Review 14(2), p.1-18.

Ruelle, D., 1993, Chance and Chaos, Penguin, London.

Shoham, Y., 1997, An Overview of Agent-Oriented Programming, in Software Agents, J.M. Bradshaw (ed.), MIT Press.

Thein, K.N.N. and Abbott, M.B., 2000, The gender issue in hydroinformatics: some pragmatic aspects, 4th International Conference Hydroinformatics 2000, Cedar Rapids, IA, USA. (Proceedings available on CD ROM only: copies available from the authors.)

Yan, H., Solomatine D.P., Velickov, S. and Abbott, M.B., 1999, Distributed environmental impact assessment using internet, Journal of hydroinformatics, 01.1, pp.59-70

Retour