Jerome Ravetz wrote Scientific Knowledge and its Social Problems (1971, 1996) and (with Silvio Funtowicz) Uncertainty and Quality in Science for Policy  (1990).  They created the NUSAP notational system and the theory of Post-Normal Science.  He is currently associated with the Institute for Science, Innovation and Society at Oxford University.

Our modern scientific view of knowledge was defined by a throwaway line in Descartes’ Discourse on Method.  Referring to his dissatisfaction with his education at school, he claimed,

“I was convinced I had advanced no further in all my attempts at learning, than the discovery at every turn of my own ignorance”. 

He was careful to say that his school was not to blame, although a little later he did a brilliant assassination job on the whole humanistic curriculum.  Readers now might not notice the irony in Descartes’ complaint.  It was not merely another case of late-adolescent angst.  For in the mention of the discovery of ignorance, his  educated readers would have recognised an echo of Socrates.  This founder of philosophy was remembered as saying that his whole life’s work was the discovery of his ignorance.  By the criteria of Socrates and all who followed, the education of the young Descartes had been a great success:  so early in life he had succeeded in discovering his ignorance!  With both Descartes and his readers knowing this background, they would recognise his complaint as the casual discarding a couple of millennia of moral philosophy.  “Know thyself” was out, “Discover truth” was in.

This point is not of merely scholarly historical interest.  The Scientific Revolution produced a variety of accounts of scientific knowledge, differing in their balance of reason and experience, and also in the strength of their claims to certainty.  But they all agreed in their tacit elimination of ignorance from their pictures of the acquisition of knowledge.  Of course, publicists for science recognise ignorance, but mainly as something out there to be conquered by the advance of science.  When scientists have undergone a lengthy and rigorous training in which they learn that for every real problem there always one (and only one) correct answer, there is little danger of them sharing Descartes’ school-leaver’s predicament.

The relevance of this issue today is, to what extent should we incorporate ignorance, as distinct from tameable uncertainty, into our reasonings about science and science policy?  I would argue that the suppression of ignorance in our debates, perhaps even its repression in our thinking, seriously impedes our management of our scientific affairs.

There is evidence that, particularly in climate science, ignorance is something of a taboo idea, even when it might seem to be most relevant.  I have two illustrative examples from the climate science area.  The first relates to a proposed scale of uncertainty, designed by James Risbey and Milind Kandlikar [1], and adopted by the IPCC [2].  This has the merit of providing a single robust scale of degrees of uncertainty, based on the notations for expressing it in numerical form.  It could be of great use in resolving the confusing variety of schemes that are employed in the various special fields that contribute to climate science.  The scale includes five degrees of increasing uncertainty, concluding with a sixth category for ignorance.  The authors were pleased to see the scale adopted by the IPCC, but then surprised to see that the category for ignorance had been deleted in the IPCC version [3].

Another example provides even stronger evidence of a consistent attitude.  Two authors who are eminent in their own fields, Sir Nicholas Stern and Leonard Smith, recently published a paper on the characterisation of uncertainty in climate science [4].  The paper is truly magisterial, bringing deep analytical clarity to this very confused subject.  But, again surprisingly, a search for ‘ignorance’ in the text produces only three citations, and two of those are incidental (p. 16 twice).  The only substantive reference relates ‘ignorance’, rather ‘recognised ignorance’, back to ‘ambiguity’ or ‘Knightian uncertainty’ (p.4).  It would seem that ignorance, in its own right as a qualitatively deeper sort of uncertainty, is not relevant here.   The absence must be deliberate, for the whole essay can be read as a detailed warning of the many pitfalls of mismanagement of uncertainty, along with the ‘fallacy of misplaced concreteness’ in relation to models. Indeed , it can be read as a Socratic exercise in all but name and vocabulary.

Particularly for that reason, I confess that I cannot agree with the absence of ignorance.  Suppose that a senior planner, responsible for the long-range defences of the Thames Estuary, approaches experts for an estimate of the sea-level rise to the end of the century.  It would be technically correct to say, “It will probably be somewhere between one and four metres, but where inbetween is a matter of ambiguity”.  The planner might prefer to be told simply, “`I don’t know,” with a review of the reasons for speculating on the likelihood of one range of values over another.

It is not as if ignorance were totally banned from policy-relevant science.  In medicine, for example, we know that we don’t know the causes of some important diseases, as indeed we are aware of our ignorance of the course of future epidemics.  The sciences do not lose public prestige because of their frankness about their deep limitations in relation to some urgent issues.  Rather, they gain trust because of their honesty with their publics.

We can see the explicit recognition of ignorance as part of the programme of a ‘technology of humility’ proposed by  Sheila Jasanoff of Harvard University [5].  It would fit particularly well with climate science, since this is after all a part of a great humanitarian project rather than a quest for profit, power or privilege.  The message of Socrates, rejected with such ultimately devastating effect by Descartes, could inform such a science and provide it with an enriching humane element.

References

[1] Risbey, J.  & M. Kandlikar, 2007: Expressions of likelihood and confidence in the IPCC uncertainty assessment process. Climatic Change, 85 (1-2), 19-31.

[2] Mastrandrea, M., C. Field, T. Stocker, O. Edenhofer, K. Ebi, D. Frame, H. Held, E. Kriegler, K. Mach, G. Plattner, G. Yohe, and F. Zwiers 2010: Guidance notes for lead authors of the IPCC fifth assessment report on consistent treatment of uncertainties, Available at https://www.ipcc.ch

[3] Risbey, J. and T. O’Kane 2011: Sources of knowledge and ignorance in climate research:  Climatic Change, 108 /4, 755-773,

[4] Leonard Smith and Nicholas Stern 2011, Uncertainty in science and its role in science policy, Phil. Trans. R. Soc. A 369, 1–24.

[5] Sheila Jasanoff 2003, Technologies of Humility:  Citizen Participation in Governing Science, Minerva 41: 223–244.