GOOD INTENTIONS ARE NOT ENOUGH

Some years ago, when I was managing a small-grants program funding research projects, I requested that the National Academy of Sciences convene a panel of experts to review the accomplishments of that program. The panel was quite positive about the program, reporting that it had contributed to the body of knowledge in the field of study, contributed to substantial development of research capacity in the developing country laboratories involved, and trained a lot of people.

I was rather surprised that, while there were some technological innovations that had been attributed to the program, these were not the highlights of the panel’s reports, but rather were seen as useful byproducts. The research program had been based on investigator-initiated, peer-reviewed proposals. The program’s funding criteria had included the likely research impact, and especially the impact on the reduction of poverty. All of the proposals had described applications and benefits to be expected from the research. The funded projects were those peer reviewers thought likely to have results that would be used.

I came to the conclusion that the Academy panelists saw this program (metaphorically) as contributing substantially to a body of knowledge, as streams contribute to a lake. Knowledge from this body would be used in many applications, as water from the lake would be used to irrigate many fields. But, as it would be difficult to ascribe the irrigation of any specific field to the contribution of any specific stream to the lake, so it would be difficult to ascribe any application of the knowledge in the research field to the results of any specific research project. They saw the value of the program not so much in terms of individual inventions, but building on the work of others in ways that others would further extend in the future.

The Academy Panel saw the body of trained people coming out of the program as very important. Not only would these people themselves do research and contribute to the body of knowledge, but they would be gatekeepers who would draw from the existing body of knowledge to create and improve applications. Combination of existing knowledge and its adaptation to local circumstances was seen as the important area of innovation, not revolutionary inventions.

In retrospect, I find the panel’s findings to be an interesting comment on intentionality. The research program was very competitive, with fewer than one in ten proposals funded. All of the funded researchers had expressed the intention that their research would be directly applied, but few of those intentions were in fact realized. We, the managers of the program, perceived it as “high risk” and believed that only a portion of the projects would have direct benefits, and that those direct benefits might often require five to ten years to realize. Our intentions too in retrospect appeared unrealistic. Yet the overall program was judged to be successful, in spite of not fulfilling the intentions neither of its managers nor its participants.

These days I spend a fair amount of time looking at ICT for Development. I find myself talking less about research and development projects, but more about other kinds of ICT projects, including:
· pilot projects, that are intended to develop new (technology based) approaches to development problems;
· demonstration projects, that are intended to demonstrate already tested approaches to new potential users; and
· best practice projects, that are intended to identify best practices related to the introduction or utilization of an innovation.
There is a lot of discussion about evaluation, especially in terms of the degree to which the projects achieve the intentions of those who funded or implemented them. There are concerns with “scale-up”, “replicability”, and “sustainability”. The terms seem very similar to those used a couple of decades ago to describe the research program, and they seem likely in the future to also prove inadequate to the reality.

If you think about it, the approach I have described is based on a teleologic metaphysical view of development and technology diffusion, a view that considers development and diffusion mainly as the implementation of intentions. That is, many people seem to assume that the development and diffusion processes should take place largely as planned, and because leaders have planned them.

There is an alternative metaphysical view, a view that I suggest may be more realistic and more useful. That view would the development and diffusion processes are primarily teleonomic. That is, the development and diffusion processes exhibit order as an emergent property of institutional processes of learning, selection, and feedback in complex systems involving large numbers of independent activities, not as the result of planning by leaders. While certainly people have intentions, and those intensions are important as the individuals determine what they will do individually, it is the impersonal institutional forces that induce a larger order in the results of millions of decisions made to achieve local objectives and based on locally available information.

The prototypical teleonomic views are the theories of the hidden hand of market economics, and of evolution in biology. The understanding that order can emerge from complexity as the result of natural processes rather than planning is only a couple of hundred years old – a very brief time for so revolutionary a concept. The controversy over Creationism versus Evolutionary Biology illustrates how very deep the divide between these metaphysical positions can be.

Lets get this back from the brink of metaphysics. NUA estimates that as of last September, there were just over 600 million people online. (http://www.nua.com/surveys/how_many_online/index.html)
This, as we all know, is the result of a very rapid dissemination of Internet technology. Most people instrumental in the creation of the Internet and the World Wide Web admit surprise in how rapidly connectivity has spread; none admits responsibility for the exclusion of 90 percent of the world’s population from connectivity at this time. There seems little alternative but to accept the idea that the dissemination of the Internet is better understood as the result of institutional processes acting on very large numbers of independent actors, than as the result of planning and foresight by these leaders.

This situation should not be surprising. If we look at previous technological revolutions, they too have this distributed nature. The Green Revolution, in which the productivity of agriculture (in a significant portion of the developing world’s farming was revolutionized by the simultaneous introduction of dwarf varieties of major food grains, chemical fertilizer, chemical pesticides, and irrigation) has been best described as the result of the decisions of millions of farmers making individual decisions to utilize the new technologies. It seems clear that most of these decisions have been based on seeing how the experiments of the early acceptors in the local communities fared with the technology. The decisions of the farmers could be influenced centrally by decisions that affected the prices of chemical inputs, the availability of irrigation, the education of farmers, and the market for the food, but the millions of independent decisions of the individual farmers were the key.

It may come as a shock to the donor agencies, but the intentions of their experts as they design pilot projects, demonstration projects, and best practice projects may have little to do with the actual rate and extent of the dissemination of information and communications technologies. Indeed, I would suggest that the expansion of the Internet connectivity or of other ICT uses and applications may not be the result of the scale-up, replication nor sustainability of the projects donor agencies design and implement. The projects can be important, but the link to dissemination of the technology may not be as simple as the project designers imagine.

More later.