I will now turn to consider some different facets or aspects of objectivity – scientific, moral, compassionate, and aesthetic. These are not different types of objectivity derived in different ways (see 1.i), but different aspects of the same objectivity encountered in different contexts. That is to say that there is a conventional distinction between these different types of objectivity, which we can continue to use provided we recognise that this does not indicate different types of justification.

The incrementality of scientific objectivity is probably already recognised by most scientists, who have to make probabilistic judgements based on the weight and quality of evidence in relation to a theory, and then meet public incomprehension of the limited evidential support for some theories that have been prematurely popularised. However, the idea that scientific objectivity is dispositional will probably require much more defending.

The traditional emphasis in Western science has been on the propositional objectivity of a theory rather than the dispositional objectivity of scientists. However, as I have already argued, the idea that propositions can represent reality does not stand up to sceptical argument (see 1.a). Moreover, the meaning of the words out of which a scientific theory is created is not purely representational: some ambiguities and differences in interpretation are unavoidable. The words of the scientific papers only secondarily convey the effects of a scientific investigation conducted with a degree of experiential adequacy. Experiential adequacy determines the success of the investigation in addressing conditions, some of which involves not just judgements about experiments and observations – including the nature and scale of testing, how to interpret the results, and how to judge the relationship of observations to theory – but also how to manage the emotional responses of all involved and how to communicate the theory and the evidence.

If we ask what makes the theory of relativity and quantum physics more objective than Newtonian physics, the answer is not that the modern theories are known to precisely represent reality in a way that Newtonian physics does not, for the historical succession of scientific paradigms has made us aware of the sceptical questions that still give uncertainty even to the latest theories that have gained acceptance. However, the scientists who pioneered these theories have been able to address conditions in a way that the Newtonian scientists have not, by identifying the gaps in Newtonian theory, having the courage to put forward and develop an alternative, and finding evidence that fulfils the predictions of the new theory. In all these respects, and possibly others, their experiential adequacy was slightly better than their predecessors’, and it is this experiential adequacy that allows them to be measured against their predecessors despite having a different theoretical paradigm.

Of course, having better experiential adequacy is rather like having a better mesh of net to catch fish in, so there is still a measure of luck in the fish that we actually catch. A scientist with a new theory may have a measure of luck in being able to find evidence for it, when another, whose experiential adequacy is just as good, fails to get the evidence she needs and is thus not taken seriously enough to support further investigation. It might be that the further investigation would, after all, have found the evidence she needed. This is an instance of where issues of presentation and emotion also become relevant to the objectivity of approach of a scientist who seeks support for further investigation. Generally, though, the experiential adequacy of the work done by scientists is the best available measure of their objectivity, because we cannot take account of the unknown factors of luck that determine their success beyond this. Objectivity does not necessarily lead to success in gaining positive results, but it will mean that a greater adequacy has been brought to bear on the investigation that leads to those results.

The objectivity of scientific research does not depend solely on the dispositional objectivity of individual scientists, though obviously these contribute. The experiential adequacy of the whole method used is also a product of the socially agreed methods used by groups of scientists and by traditions of scientific practice. This socially agreed adequacy is also dispositional, but the dispositions are those of a group of people acting together. The success of this social level of investigation depends on its degree of integration (see 6.f on social integration), including the coherence of the approach taken by each member of a group, but also the extent to which they collectively recognise the possibility of error (see 5.c & d).

Philosophers of science have long been questing for an account of the objectivity of science, but have failed to find one. To avoid either naive realism or a scientific relativism in which Aristotle’s science is as good as Einstein’s, we need to consider the history of successful practice in science without these extremes of metaphysical assumption. The two great figures who have tried to do this, however, Lakatos and Kuhn, both failed to reach an account of the objectivity of scientific theory. Both, however, came close to an account of objectivity in the skills of scientists, limited only from drawing fuller conclusions by their assumption that an account of objectivity could not be psychological, because they believed that this would be irredeemably ‘subjective’. Kuhn described these skills and attitudes of scientists as ‘puzzle-solving ability’:

Taken as a group or in groups, practitioners of the developed sciences are, I have argued, fundamentally puzzle-solvers. Though the values that they deploy at times of theory-choice derive from other aspects of their work as well, the demonstrated ability to set up and to solve puzzles presented by nature is, in case of value conflict, the dominant criterion for most members of a scientific group. Like any other value, puzzle-solving ability proves equivocal in application. Two men who share it may nevertheless differ in the judgements they draw from its use. But the behaviour of a community which makes it pre-eminent will be very different from that of one which does not. ^[1]

Without such a detailed survey of scientific success here, I can only point to the implicit support offered by the conclusions of Lakatos and Kuhn to the thesis that scientific objectivity consists in the qualities of scientists and their communities. The response to a puzzle depends most basically on the recognition that existing answers are inadequate, and on a continued provisionality in handling possible alternative answers. The need to be able to keep working within a given paradigm, maintaining a justified confidence in it so long as it is fruitful, needs to be balanced with the critical skills involved in deciding whether to move to a different paradigm. The successful scientist has to follow the Middle Way in yielding neither to a dogmatic relativism that would lead her to drop a successful paradigm too soon, nor to a dogmatic attachment to the current paradigm that stops her from treating it critically. The same kind of balancing of judgement also applies to smaller decisions where a particular line of enquiry needs to be continued or dropped within a particular paradigm[2].

One particular scientific issue which illustrates the role of dispositional objectivity in science especially well is that of publication bias. Scientists have an obvious interest in being able to highlight positive discoveries, and are thus very likely to give these maximum publication exposure, whilst negative results often remain unpublished[3]. Given that negative results may in the long run provide more useful information to guide future theorisation and research than positive ones, this tendency is very unhelpful to the objectivity of science. In 1998 a review found that in the entire medical research taking place in China, no negative findings had been published[4]. One could conclude from this with reasonable justification that medical research in China is less objective than in other countries where at least some negative results are published, not because the theories considered by Chinese medical researchers are necessarily further away from reality, but because the conditions for finding out incrementally how close they may be to reality are not present. Of course this defect in the objectivity of scientific tradition in China may in some cases be compensated by individual objectivity, but the conditions make this less likely.

In parallel to scientific objectivity where the organised social pursuit of knowledge takes place, there is also an individual factual objectivity. This is the degree of objectivity with which a given individual pursues what they perceive to be facts about the universe. Exactly the same criteria apply, apart from a different relationship to the social aspect of scientific objectivity. The individual may still be dependent on the objectivity of organised science to support individual objectivity. In the modern world, the vast majority of our factual beliefs reach us through the education system and the media and are thus social in origin, but we nevertheless need objectivity in assessing the reliability of these sources. Instead of scientific theories investigated directly, the individual may well have implicit theories such as “The BBC is a good unbiased source of political information.”

We may be made aware by examples of public ignorance (e.g. the majority of Americans rejecting Darwinism and anthropogenic global warming) that a relatively objective science is not sufficient to ensure a population that is objective to the same degree, and the objectivity of science is ultimately dependent on that of the wider population that supports it. I would suggest that individual factual objectivity has been relatively neglected, that factual investigation has been left too much to specialists, and that the influence of the paradigm of knowledge (as opposed to mere ‘opinion’) rather than justified provisional belief has contributed to this.

---

[1] Kuhn (1996) p.205

[2] See Ellis (2001) 2.b for a more detailed discussion

[3] Goldacre (2008) pp.212-16

[4] Vickers et al (1998)