Autonomy and Objectivity of Science

Abstract

This article deals with the problematic concepts of the rational and the social, which have been typically seen as dichotomous in the history and philosophy of science literature. I argue that this view is mistaken and that the social can be seen as something that enables rationality in science, and further, that a scientific community as well as an individual can be taken as an epistemic subject. Furthermore, I consider how scientific communities could be seen as freely acting and choosing agents. Fundamentally, this boils down to the question whether we accept the voluntarist conception of human beings, one consequence of which is that scientists possess, in principle, the capacity for deliberative reflection and choice. If this is accepted, we can talk about the degrees of autonomy that communities possess. I also examine what kinds of decisions an autonomous community should make in order to produce objective knowledge. My suggestion is that objectivity be understood as intersubjectivity: a view is objective when it has been exposed to critical reflection from various points of view, and due to this, transcends subjective idiosyncrasies.

Acknowledgements

I would like to thank Bart Karstens for his extremely useful comments and suggestions on this paper. This work was supported by NWO, the Netherlands Organisation for Scientific Research.

Notes

More precisely, these are the words that appear in the proceedings of the 1965 Bedford College colloquium (Lakatos 1970, 178).

There are grounds to say that even Kuhn was rather internalist after all. He repeatedly emphasized the role of epistemic factors in theory decision although they did not work as an algorithm to determine a uniquely best theory (e.g. Kuhn 1970, 152–155, 199).

Schmitt divides social epistemology and its areas of investigation into three branches: the role of social factors in individual knowledge, the organization of the cognitive labour of individuals and groups of individuals, and the nature of collective knowledge. Although all these branches study social aspects of knowledge, the first two are committed to epistemological individualism, i.e. the individual is the central epistemic subject on which social factors have an influence (Schmitt 1994, 4; Solomon 1994; cf. Kusch 2010).

An example would be the following network of appeals to the authority: A knows that m; B knows that n; C knows (1) that A knows that m, and (2) that if m, then o; D knows (1) that B knows that n, (2) that C knows that o, and (3) that if n and o, then p; E knows that D knows that p (Hardwig 1985, 348).

Oreskes (2008, 255) claims that there are some historical and sociological ‘prize-winning books by leading academicians’ that overcome cognitive individualism and explicate the role of the social in producing the rational.

Kusch is another example of a social epistemologist who wishes to go beyond classical epistemological individualism. Kusch calls ‘naturalistic’ the type of social epistemology that challenges individualism in light of the results of natural and social sciences, pointing out that either our best scientific theories attribute knowledge to groups or the understanding of contemporary science is possible only if individualistic intuitions are dismissed. By contrast, Goldman dubs this type of introduction of collective believers and knowers as ‘expansionist social epistemology’. See Kusch (2010) for further references on either of these orientations or on social epistemology including Goldman in general.

There is naturally an extensive literature on the concept of rationality into which we cannot delve in this paper. Nevertheless, I believe Laudan's ‘instrumental rationality’ can be used to illustrate the issue here, in particular because it appears very similar to Solomon's consequentalist sense of rationality. Laudan's idea is that methodological rules are hypothetical imperatives, designed to be means to cognitive ends or tools for performing a task. Thus, the rules of the form ‘one ought to do x’ should be understood as ‘if one's goal is y, then one ought to do x’. Rationality is thus acting in a way that promotes one's ends. See Laudan (1984, 34, 41; 1987, 21, 24, 227). Then, even if ‘non-empirical decision vectors’ have shaped the practice of science, they don't seem to be factors that make achieving one's goal more likely if that goal is empirical success. In other words, one would hardly incorporate them into one's theory of rationality, stating for example that ‘assess the salience of data and make your theories agree with scripture, if you aim at empirical success’. Further, Richardson wonders what the role of non-empirical factors is in Solomon's theory, since it is empirical success that matters in decisions between different theories. He asks, whether our normative theories on rational consent in science should be based on empirical success alone (Richardson 2008, 252; similarly Longino 2008, 244–245). Moreover, Clough (2008) remarks that the distinction between empirical and non-empirical categories appears to reintroduce the internal/external polarity in science studies, although of course this would not renew the individual/social dichotomy, because the rational/irrational distinction could be made on the social level. Clough's suggestion to use relevant/irrelevant categories is nevertheless valuable, because this allows that any factor, vector, or value can become a (relevant) reason in theory decisions. Solomon appears to agree with Clough (Solomon 2008, 284). Clough's suggestions also supports the central idea in this paper that it is the process that amounts to justification and objectivity in science, and not so much an issue of what kind of factors support a theory. As she says in reference to Anderson (2004), ‘value judgments can be shown to be amenable to reflective deliberation’ and (we might add) become reasons for theory's acceptance or rejection as a result (Clough 2008, 273).

Barnes (1999) writes that although it is difficult to find a systematic reflection on anti-individualistic accounts of ‘choice’ and ‘agency’, the central problems of sociology are those of collective agency. His suggestion is a version of sociological compatibilism, according to which it is an inherent sociability that makes voluntaristic notions applicable and that makes it possible for human beings to affect each other through the medium of this kind of discourse. Alternatively expressed, it is in virtue of being mutually accountable and susceptible as social creatures that we can act free and voluntarily (Barnes 1999, x–xi, 73, 143). Barnes's sociological compatibilism appears compatible with the account of the autonomy of scientific communities developed in this article.

For further discussion, see Kuukkanen (2011).

For some references, see note 12 below.

Oreskes (2008) argues that no theories in the history of science have met Solomon's standard of appropriate consensus either. Interestingly, Solomon (2008) replies that she is not disturbed by this, because her social empiricism offers a framework within which one can distinguish the scientifically productive dissent from unproductive.

There is a debate in philosophy of mind on whether determinism and free will are compatible. While the common sense view is that determinism is not compatible with the freedom of the will, many philosophers nevertheless believe that it is (Ayer 1954; Frankfurt 1982; Watson 2003), not because in some cases there are no causes for actions but because of the absence of compulsion or constraint. See note 8 above on Barnes's sociological compatibilism.

Cf. Goldman (1994, 283) on the notion of institutional goal and its relation to normativity.

Kusch (2010) writes (in reference to Craig 1990) that according to genealogical social epistemology and historical epistemology we move from ‘protoknowledge’ of primitive epistemic communities to our concept of knowledge via the process of ‘objectivization’. One requirement in the process is that ‘the perspectival or indexical character of protoknowledge is weakened’. This seems to be an expression of the same idea that the (cultural or individual) idiosyncratic elements in our beliefs gradually decrease through intersubjective scrutiny.