How Should We Approach the Question What is Science?
There are activities whose criteria of success are public and explicit. That is so for games with standard rules, and with building projects under contract, and of course for political campaigns which need to end with a clear winner.
But when we turn from specific, local, clearly specified activities of this sort what counts as success tends to be vague and subject to ambiguity or controversy, often enough at odds with what is publically avowed as the aim (think of the Iraq war!).What about science? That is a very broad, and not clearly defined, area of human enterprise. Specifically, attempts at demarcation, at drawing lines between what is (genuine) science and what is not have been famously unsuccessful. Yet we have on the whole a fairly confident sense of our ability to refer to, discuss, admire, and argue over science and the sciences. Even if the concept of science is a cluster concept, it seems to be accepted that its central activities include the construction of theories and models, and that questions of success and failure are by and large settled within the relevant scientific communities themselves.
But this is just where philosophical battle lines are drawn: what are the criteria of success and failure? Is there a ‘bottom line', a basic criterion of success?
Scientific realism, in its various versions, tends to focus on truth, or on criteria that involve truth, such as true explanation, or accurate representation. Empiricism, in its various versions, tends to focus on truth ‘within our ken', empirical adequacy. These are contrasting assertions about what counts, at bottom, as success in science. It is not assertions, and variations thereon, that are lacking, but ways to judge these assertions.
There is a serious, relevant inquiry that has not been taken on to any great extent as yet:
How can we determine what are the criteria of adequacy in a given human activity?
And specifically, how can we assess such a claim as that in (current, or Western, or...) scientific practice the ‘bottom line' criterion of adequacy is empirical success rather than truth overall, or the other way around?
Careful examination and analysis of science in practice is what is needed here, but the analysis needs to pursue a question that is not peculiar to any specific episode in scientific practice.
While I hope that this problem will be taken by the horns in new research, I can for the moment point to two inquiries which seem relevant.
The first is Frederick Suppe's paper on credentialing, and the second is the attempt by Clark Glymour to follow Hermann Weyl's version of ‘co-ordination', which I have tried to re-codify as empirical grounding.
5.1 Suppe on Credentialing
Fredrick Suppe launched a sustained effort to separate two senses of “confirmation” that had been conflated in the philosophical literature: belief formation and evidential support.
The topic of belief formation, based on evidence and reasons of all sorts, is the traditional subject of the supposed logic of induction or abduction and Bayesian confirmation theory. Evidential support is what we find in actual scientific publications when data, and auxiliary information, are submitted to back up scientific claims. To the traditional eye, the two must be the same, and Suppe's radical assertion, on the basis of a thorough examination of the form of articles in scientific technical journals, is that they are not the same at all.
What is significant especially in Suppe's writing on this topic is that it presents a quite new characterization of what is sought and pursued as success in the scientific reports of data, experimental and observational findings, and the conclusions based thereon. The criterion for new scientific claims to satisfy is that they should come with satisfactory credentials to enter the body of scientifically accepted information and theory.
It is unfortunate that Suppe presents his case enmeshed in much extraneous pleading, argument, and debate (for example, with social constructivists). But the details of his case can be gleaned from three lengthy papers (Suppe 1993, 1997, 1998). Specific analyses of technical literature and episodes in the recent history of science are presented there, to show how the actual narrative and argumentative structure differs from any of the schemata presented in traditional and formal epistemology.
The process of credentialing actually found is basically a ‘mine sweeping' operation to remove obstacles and possible objections or doubts that could reasonably be raised against the claims submitted by the study's authors. The typical paper reporting experimental studies:
• presents the reduced data or results of the experiment
• details the relevance of the experiment and its results for the target scientific community
• provides details concerning the experimental set-up, the apparatus, and the circumstances of the experiment that would be needed to replicate or evaluate the study
• provides an interpretation of the reduced data which yields the specific experimental claims
• marshals evidence to anticipate and remove specific doubts that the data could have been due to interference, inadequate statistical analysis, or be ‘artifacts of measurement'
• marshals further arguments to rebut possible alternative interpretations of the data.
The refutation of envisaged alternatives does not, of course, imply the truth of the claim unless the range of alternatives is logically exhaustive, which is not possible in practice. But it does establish the credentials that the claims in question must have to be candidates for acceptance within the target scientific community. Suppes remarks:
To the extent that there are no unrebutted objections or competing interpretations, the typical form of the interpretative argument is “to the Best Explanation. Second Edition. London: Routledge. Musgrave, Alan “Strict Empiricism versus Explanation in Science” pp. 71-94.
Papineau, David (1996) The Philosophy of Science. Oxford: Oxford University Press.
Piatelli Palmerini, Massimo (1984) Livelli di Realta. Milan: Feltrinelli Editore.
Suppe, Frederick (1993) “Credentialing Scientific Claims”, Perspectives on Science 1, 153-203.
Suppe, Frederick (1997) “Science Without Induction”, pp. 386-429 in John Earman and John D.
Norton, Cosmos of Science: Philosophical Problems of the Internal and External Worlds.
Pittsburgh: University of Pittsburgh Press, 1997.Suppe, Frederick (1998) “The Structure of a Scientific Paper”, Philosophy of Science 65, 381-405. van Fraassen, Bas C. (1976) “To Save The Phenomena”, Journal of Philosophy 73: 623-632. van Fraassen, Bas C. (1980) The Scientific Image. Oxford: Oxford University Press.
van Fraassen, Bas C. (1989) Laws and Symmetry. Oxford: Oxford University Press.
van Fraassen, Bas C. (2005) “Wouldn’t It Be Lovely? Explanation and Scientific Realism”, Review Symposium for Lipton, Inference to the Best Explanation, 2nd ed. Metascience14: 344-352.
van Fraassen, Bas C. (2007) “Reply: From a view of science to a new empiricism”, pp. 337-383 in Monton, Bradley (ed.), Images of Empiricism: Essays on Science and Stances, with a Reply from Bas van Fraassen. Oxford: Oxford University Press, 2007.
van Fraassen, Bas C. (2008) Scientific Representation: Paradoxes of Perspective. Oxford: Oxford University Press.
van Fraassen, Bas C. (2012) “Modeling and measurement: the criterion of empirical grounding”, Philosophy of Science 79: 773-784.
van Fraassen, Bas C. (2014) “The Criterion of Empirical Grounding in the Sciences”, pp. 79-100 in Gonzalez, W. J. (ed), Bas van Fraassen’s Approach to Representation and Models in Science. Dordrecht: Springer Verlag.
Weyl, Hermann (1927/1963) Philosophy of Mathematics and Natural Science. NewYork: Atheneum.