Data, development and democracy: What is missing?

In an opening address to a UN colloquium on science, technology and society held in 1979, past president of the Royal Netherlands Academy of Arts and Sciences Dr Casimir said, "Science and technology cannot be applied to development. Science and technology are an essential part of development. One does not apply one's lungs to respiration, nor one's heart to the circulation of blood nor one's legs to walking. If we regard science and technology as a crutch, it will at best provide a halting gait. If we regard them as a transplanted heart, it will sooner or later be rejected by the receiver." By saying this Casimir did not disparage science rather he tried to differentiate between 'science for development' and 'science as development'. When we put forward our argument about super critical technology in a coal-fired power plant, we should not think of applying it to our development rather it becomes a part of the development itself. So, measure of its efficacy should be the data it generates. But data is neither good, nor bad, nor is it neutral. Data does not speak for itself. It must be analyzed and interpreted. Data can be socially beneficial, but data can morally bankrupt us. Governments can use data to improve governance or to centralize control. People can use data to empower themselves and their communities or to reify injustices and inequalities.
Cathy O' Neil, in his book, "Weapons of Math Destruction", argues that we live in the age of the algorithm that process data. Increasingly the decisions that affect our lives such as how much to pay for insurance are being made by mathematical model. But these models are opaque, unregulated and not contestable. This creates, in her word, "a toxic cocktail for democracy".
Generally speaking, science is not done by popular vote. The backbone of science is data and information. The access and the control of information can transform citizenship. Information can be controlled by an interpretation in order legitimise certain political option. Political parties are an essential component of democracy. They mobilize citizens behind a particular vision of society. However, in many countries political parties fail to respond to citizen's concern and entire democratic process suffers. In a sustainable democracy political will should deeply entrenched in the fabric of society. Big data or small data, both can play a part in social segregation, social control, political will and judgement. This is a new challenge for democracy in the modern information age.
Science also sometimes contaminated by feelings. As an example, couple of years ago, in a panel discussion about string theory in Toronto, a Stanford professor invoked his executive privileges. He asked the audience members for a vote on whether, by the year 3000, the value of the cosmological constant would be explained by the anthropic principle or by fundamental physics. The panel split 4 to 4, with abstentions, but the audience, who are also scientists, voted overwhelmingly for the latter possibility. Panel chair concluded, "We have made some progress in sharing our feelings." However, this feeling is rooted in a scientific process as delineated in physics.
Between data and democracy we have development process. It is a sandwich. When we consider any new development, data and democracy play a vital part.  Former Secretary General of UNCSTD Joao Frank da Costa spelt out some new understanding about development, which he called "Twelve must for development". These are as follows: (1) Development must be total. This means that the development is not limited to economic factors. (2) Development must be original. Here he mentioned about the diverse nature of development style (3) Development must be self determined. (4) Development must be self generated. Here horizontal cooperation with other developing countries in order to achieve self reliance is the main focus. (5) Development must be integrated. This means that industrial and agricultural sectors must be developed jointly together with the system for education and training. (6) Development must respect the integrity of the development. This is self explanatory. However, both natural and cultural issues need to taken into account. (7) Development must be planned. This means that economic development cannot ensure the equitable diffusion of science and technological potential. Therefore, a planned approach is essential. (8) Development must be directed towards a just and equitable social order. Here benefit of all sectors of the population from a development has been considered. (9) Development must not insulate less developed regions into reservation. (10) Development must be innovative. This means that it should not depend on some sort of imported technology or outmoded technologies. (11) Development planning must be based on a realistic definition of national needs. And last but not least (12) Development must be democratic. This means that society's goal is not all scientific and technological. This must not be allowed to assume control.
How data play a part in a development concept? I can spell out one above mentioned 'must' for development. In order to make a development 'total', we need data related not only to economic factors, but also to both social and cultural factors. Think about how destruction of rainforest is affecting the livelihood of Amazonians. Some development requires environmental impact assessment and also socioeconomic or health impact assessment. These require objective assessment of data using collective tools, technologies and processes, which are known as 'datafication'.
Bertrand Russell in his essay 'Scientific Government' argued that a government is scientific does not mean that it is composed of men of science. Napoleon's government was full of scientists including Laplace. Laplace was proved to be incompetent in the eyes of Napoleon's government and dismissed in a very short period of his appointment. This, as Russell argues, does not mean that a government is not scientific or scientific rather Russell puts it, 'I should define a government as in a greater or less degree scientific in proportion as it can produce intended results'.  Russell further mentioned that 'the greater the number of results that it can both intend and produce, the more scientific it is.' Russell also argued that owing to the increase of knowledge, it is possible for governments to achieve more intended results than it was before. We now encounter big data. This data can increase our knowledge and understanding. Obviously it depends on how we interpret the data.  So, question is-is intention data driven or data interpretation intention driven? It could go both way and create some problem in decision making.
Our mathematical modellers who handle the data should take more responsibility for their algorithms and policy makers should take responsibility for regulating their use. There are plenty of alternatives in the world. What missing is an alternative thinking of alternatives. We should turn our political or social will towards that kind of thinking. This will foster citizenship and democracy.r
Dr Kanan Purkayastha teaches chemistry and environmental sciences in United Kingdom. He is a fellow of the Institution of Environmental Sciences in UK