Avoiding another Climategate: For a democratization of science

The leaked emails from the Climate Research Unit (CRU) in Essex have led to highly controversial debates on the monopolization of data and scientific research. This "Climategate" has not only culminated in Phil Jones stepping back from his position as the director of CRU, but has given skeptics of global warming (I use "skeptics" as a euphemism) arguments that undermine the efforts at Copenhagen.

According to skeptics, the emails between leading climatologists show collusion to deliberately surpress and bias scientific information. The statistical "trick" applied in the questioned paper by Mann, Bradley and Hughes (1998) is often cited as proof of obvious data manipulation. In contrast, advocates of action on global warming argue that these passages are quoted out of context, thus distorting the intended meaning. Even if the emails reveal some unscientific "tricks", the advocates (and with them many scientists) argue that these little glitches by far do not weaken or falsify the massive evidence for climate change.

While this "Climategate" has no serious implications for the evidence on climate change, it does well illustrate the power of figures and the political dimension of science in general: Data, are - unlike the Latin origin would suggest - not simply given but need to be measured, cleaned and interpreted. Unless there is full transparency on the exact procedures applied, we will never know how reliable the information actually is. And even if all research is correct, the choice on which findings are allowed to diffuse and spread (i.e. released in major scientific journals) raises many issues.

The CRU is one of the few institutions that manage the data used in most of the major climate models; yet, most of the data is not easily available to the public. No wonder that this monopolization of knowledge gives rise to (possibly justified) critique on data manipulation: If there is no way for a mortal to verify the data, why should we not question the quality of it?

The CRU is not the only example for monopolization of knowledge. As another example, most of the comparative economic indicators and indices are composed and authored by the World Bank, which more often than not do a good job in obfuscating the underlying methodology. Journals, key institutions in diffusing scientific knowledge, are often found to be biased and uncritical (particularly in pseudo "natural sciences" such as economics). If data is requested for replication purposes, scientists are often reluctant to hand out the data set, pointing out that the data has already been verified by the Journals. This uncooperative rent seeking does not only undermine the credibility of science, but also slows down progress by re-inventing the wheel through time consuming reverse engineering.

In order to save the credibility and true nature of science (what I believe lies in an open and reasoned discourse), a radical democratization of knowledge is therefore needed: Democracy, not in the sense that scientific facts are subject to the voice of the masses, but rather democracy in the sense of openess and participation. Knowledge must be available to everyone to raise capabilities of verification and accountability. It might need a PhD to understand the data, but this does not imply that the data should not be available for everyone to download. Web 2.0 projects such as Wikipedia or decentralized/distributed computing reveal the utility and power of the people.

Some institutions are leading such a process of democratization. The CID at Harvard, for instance, publishes the datasets of its respective papers. In fact, more and more social scientists (particularly economists) have begun to upload datasets, including program codes. MIT's OpenCourseWare offers a variety of lecture notes, exams and podcasts for download. In response to the "Climategate", the UK Met has now likewise announced to publish some of its data.

Certainly, releasing data could give rise to problems related to innovation and patent rights: I agree that scientists who spent large efforts on collecting data are rather reluctant to freely distribute their valuable results. However, binding agreements that allow the scientist to reap the monopoly rents for a certain time period could easily resolve the conflict. Only if scientific institutions dismantle the elitist facade and make knowledge available for everyone, will science be able to remain her deserved credibility and withstand silly accusations by vested interests.

GX.