Abstract
The introduction (a) defines robustness and solidity; (b) provides a systematic analysis of the structure of the problem of robustness; (c) stresses several important difficulties, makes suggestions intended to help to overcome them, and points to issues waiting for further work; (d) sketches the philosophical implications related to the solidity problem; (e) gives an overview of the different chapters of the present book.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Thanks to William Wimsatt for offering to re-print this fundamental chapter in the present book. It is all the more important because this chapter is a pivotal source of inspiration for most of the contributions of this book, and is very often quoted or exploited. So the reader will have the original chapter at hand.
- 2.
Several authors of this book have adopted my terminological proposal, but of course, alternative terminological options are possible, as soon as they are clearly specified. For example Mieke Boon, in her paper, chooses to retain the expression “robustness notions” to encompass all the various uses connected to the intuitive idea of robustness. Then she decomposes this umbrella expression into different but related features (such as reality, reproducibility, stability etc.) that can be attributed to specified kinds of things (independent reality, physical occurrences, observable and theoretical objects, etc.), features that may differ in status (metaphysical, ontological, epistemological, etc.). In such a framework robustness is a generic notion, and we then have different species of robustness (robustness in the sense of reliability, robustness in the sense of multiple determinations, etc.).
- 3.
- 4.
The episode of the ‘discovery of the weak neutral currents’ on which I rely in my paper would also be a good example. See notably the quotations of scientists given in Schindler (201X).
- 5.
Admittedly, this is not a typical case of robustness, because as soon as real experiments are conceived as improvements on one and the same experiment, their independence is obviously discussable, and in any case they are not independent in the same sense as multiple experiments of different kinds. (For more details on this point, see below Section 1.9.2 the presentation of Dufour’s contribution.) But as we will see, the independence is always problematic, and, anyway, the configuration involved in Dufour’s case study is a widespread situation that must be considered in relation to the robustness issue.
- 6.
Bayesians might contest the claim that the estimations of the arrow strengths are indeed “more qualitative”, on the basis that they are able to quantify the weight of each arrow through numerical measures of evidential support. (I thank Stegenga to have drawn my attention to this point.) But, to my eyes, such kinds of attempts are at least useless with respect to the aim of characterizing real, ongoing scientific practices, and sometimes pernicious with respect to the realistic pretention of the numerical values they put forward and the algorithmic-transparent model of rationality they suggest. (see below, the end of Section 1.4, and Chapter 10, especially the conclusion.) So, at the end of the day, I think that robustness is indeed a qualitative notion (as is any human judgment), even if quantified modelisation can sometimes be clarifying. As Stegenga concludes, after having noticed that “Philosophers have long wished to quantify the degree of support that evidence provides to a hypothesis” and having regretted that in such context they “developed confirmation theory ‘given a body of evidence e’, without worrying about what constitutes a ‘body of evidence’ ”: “at best, the problem of discordance suggests that robustness is limited to a qualitative notion.” I agree with Stegenga’s conclusion, perhaps more strongly than he does himself, given his cautious formulation. In any case, the study of real, ongoing practices clearly shows that different practitioners very often weight differently the multiple derivations for and against a given hypothesis in a historical situation (see for example Dufour’s paper as an illustration among others). So if weights could be meaningfully attributed to derivations, they would likely be different from one scientist to another one – which is of course not in the ‘objectivist’ spirit of the Bayesianist enterprise.
- 7.
For further developments on this issue, see also Section 1.5, third and fourth points.
- 8.
In Section 1.8, we will see that the variable ‘number of derivations’ is no more independent of the variable ‘degree of independence’ than of the variable ‘strength of the derivation’.
- 9.
This point is closely related to what has been developed above, Section 1.3, about the issue of the weights associated with the derivations. An alleged negative argument once proposed in the history of science against an X but taken by a given analyst to be very weak if not totally dismissed as an argument (equated with something that counts for nothing, or if one prefers, associated with a weight equal to zero), will not appear at all in the robustness scheme that this analysis proposes as a reconstitution of the historical situation.
- 10.
Some might claim that images or maps are reducible to complex networks of propositions and hence do not correspond to a different type of node than in the preceding case. This might be a controversial point. In their paper, Allamel-Raffin and Gangloff argue that images are not reducible to propositions and that they play a specific role in scientific argumentation. I am inclined to think they are right. Anyway, it is not my problem here to discuss this point. My intention is only to illustrate the diversity of the scientific items that can possibly constitute a node in a solidity scheme.
- 11.
Note that the same terms were already employed in the previously considered case, ‘X = a propositional entity’. Actually, these terms, and especially the reliability vocabulary, have a very broad scope and can be applied to propositional results as well as to procedures. (In this respect, the lexicon of ‘reliable’ could have been a good alternative to the vocabulary of ‘solidity’). But note also that the reverse does not hold: all the terms used to name the solidity of a propositional result do not automatically apply to procedural achievements. In particular, talk of ‘truth’, ‘objectivity’ and what is ‘real’ are restricted to propositional entities. They are not, in ordinary usage at least, employed to pick out procedures, methods, derivations and the like. These differences at the level of the vocabularies reflect a difference of epistemic kinds.
- 12.
This is of course just a thought experiment in order to show how the solidity scheme works, not a positive claim about the possibility of ‘measuring’ the solidity values that in fact hold or should have held in a given empirical situation. As already suggested in note 6, this possibility is, to say the least, highly problematic.
- 13.
The quotes are intended to prevent the reader from equating this ‘initial’ to a given moment of the history of science at which one would have succeeded in measuring the actual, objective solidity values of some existing scientific derivations. This is not at all the sense of the present exercise. The ‘initial’ and the ‘final’ name two logical moments of the proposed thought experiment. The thought experiment imposes by fiat such and such solidity values to certain elements of the robustness scheme (which are the ‘initial’ values) and then discusses, on this basis, what consequences result ‘next’ (i.e., what are the ‘final’ solidity values and how they result from the ‘initial’ state). Of course, this thought experiment is intended to help us to understand what happens in real historical situations. But this does not imply that in real historical situations anyone is able to associate explicit objective measures to scientific derivations. The relation with the history of science, and the way the thought experiment is used to understand real situations, will be sketched below Sections 1.7.2 and 1.7.3.
- 14.
Pickering’s symbioses apply to real time dynamic scientific practices and their various (material, intellectual and social) ingredients, rather than to elements of an idealized and ‘synchronic’ robustness scheme as is the case in the reasoning just above. For more about the symbiotic conception of scientific development and references, see Section 1.7.3.
- 15.
Thomas Kuhn introduces the idea of “local holism” in a 1983 paper in which he tries to understand the incommensurability of scientific theories in terms of the (locally) different linguistic structures of these theories (On Kuhn, robustness and incommensurability, understood in a structural framework inspired by network theories, see Nickles’ paper, Chapter 14.) To decide how local is this local holism (and hence how ‘extended’ or ‘spread’ is the incommensurability), or to decide the size of the network relevant to robustness assessments, is the same kind of problem and leads us to back to the tension between holism and modularity stressed Section 1.2.
- 16.
- 17.
This is actually a very difficult task. Hacking made an attempt in this direction in Hacking (1992). He distinguishes three main categories (“things”, “ideas” and “marks”) and fifteen elements distributed on them (See the quotation note 21 below for examples of these elements. See also Mieke Boon’s paper, Section 1.2, for a presentation and brief discussion of Hacking’s typology). More generally, and as already noted, Boon’s paper itself aims at building a typology of the ingredients involved in scientific practices that might be candidate to robustness attributions. Further work on this issue is, in my opinion, strongly needed. Clearly, the general – highly criticized but still in use – distinctions such as cognitive/social and internal/external factors are completely non-operative for the analysis of detailed case studies.
- 18.
- 19.
The degree of complexity and the opacity of the configurations could also be mentioned here, but I will leave these aspects aside.
- 20.
In his contribution to this book, Pickering also considers (and primarily focuses on) the ontological side of robustness (see below, Section 1.9.5).
- 21.
“Let us extend Duhem’s thesis to the entire set of elements (1)–(15). Since these are different in kind, they are plastic resources (Pickering’s expression) in different ways. We can (1) change questions, more commonly we modify them in mid-experiment. Data (11) can be abandoned or selected without fraud; we consider data secure when we can interpret them in the light of, among other things, systematic theory (3). (…) Data assessing is embarrassingly plastic. That has been long familiar to students of statistical inference in the case of data assessment and reduction, (12) and (13). (…) Data analysis is plastic in itself; in addition any change in topical hypotheses (4) or modelling of the apparatus (5) will lead to the introduction of new programs of data analysis” (Hacking 1992, 54). “Far from rejecting Popperian orthodoxy, we build on it, increasing our vision of things that can be ‘refuted’ ” (Hacking 1992, 50). “The truth is that there is a play between theory and observation, but that is a miserly quarter-truth. There is a play between many things: data, theory, experiment, phenomenology, equipment, data processing” (Hacking 1992, 55).
- 22.
This is what happened historically. For references about this case study, see my contribution to this volume, in particular notes 1 and 6.
- 23.
See Hacking (1999, 2000). For further consideration of the inevitabilitist/contingentist issue and relevant references see Soler (2008b) (for the presentation of the problem and its situation in the landscape of the Science Studies) and Soler (2008c) (for an analysis of the structure of the problem and its internal difficulties). See also Soler (201X) for a general argument in favor of contingentism. In french, see Soler (2006a).
- 24.
- 25.
For problems of individuation that arise here, see Chapter 9.
- 26.
More generally, students of history should be curious why, even in wider contexts, such as ‘proofs’ of the existence of God, thinkers thought it necessary to provide more than one proof, each one supposedly decisive in itself.
- 27.
In this respect, see Van Bendegem et al. (2012). Krömer’s conclusion just above exactly expresses the spirit in which this book, entitled From Practice to Results in Logic and Mathematics, has been impulsed, following a Conference organized by the PratiScienS group and myself in June 2010. See http://poincare.univ-nancy2.fr/PratiScienS/Activites/?contentId=6987.
- 28.
This chapter is here presented by Thomas Nickles.
References
Hacking, Ian. 1992. “The Self-Vindication of the Laboratory Sciences.” In Science as Practice and Culture, edited by A. Pickering, 29–64. Chicago and London: The University of Chicago Press.
Hacking, Ian. 1999. The Social Construction of What? Cambridge, MA: Harvard University Press.
Hacking, Ian. 2000. “How Inevitable are the Results of Successful Science?” Philosophy of Science 67:58–71.
Kuhn, Thomas. 1983. “Commensurability, Comparability, Communicability.” In Proceedings of the 1982 Biennal Meeting of the Philosophy of Science Association, edited by P.D. Asquith and T. Nickles, 669–88. East Lansing, MI: Philosophy of Science Association.
Nederbragt, Hubertus. 2003. “Strategies to Improve the Reliability of a Theory: The Experiment of Bacterial Invasion into Cultured Epithelial Cells.” Studies in History and Philosophy of Biological and Biomedical Sciences 34:593–614.
Perrow, Charles. 1984. Normal Accidents: Living with High-Risk Technologies. New York: Basic Books.
Pickering, Andrew. 1984. Constructing Quarks, a Sociological History of Particle Physics. Chicago and London: The University of Chicago Press.
Pickering, Andrew. 1995. The Mangle of Practice: Time, Agency and Science. Chicago and London: The University of Chicago Press.
Pickering, Andrew. 201X. “Science, Contingency and Ontology.” In Science as It Could Have Been. Discussing the Contingent/Inevitable Aspects of Scientific Practices, edited by L. Soler, E. Trizio, and A. Pickering. In progress.
Schindler, Samuel. 201X. Weak Neutral Currents Revisited (under review).
Soler, Léna. 2006a. “Contingence ou inévitabilité des résultats de notre science?” Philosophiques 33(2):363–78.
Soler, Léna. 2006b. “Une nouvelle forme d’incommensurabilité en philosophie des sciences?” Revue philosophique de Louvain 104(3):554–80.
Soler, Léna. 2008a. “The Incommensurability of Experimental Practices: The Incommensurability of What? An Incommensurability of the Third-Type?” In Rethinking Scientific Change and Theory Comparison. Stabilities, Ruptures, Incommensurabilities? edited by L. Soler, H. Sankey, and P. Hoyningen, 299–340. Dordrecht: Springer, Boston Studies for Philosophy of Science.
Soler, Léna. 2008b. “Are the Results of our Science Contingent or Inevitable? Introduction of a Symposium Devoted to the Contingency Issue.” Studies in History and Philosophy of Science 39:221–29. Dordrecht: Springer, Boston Studies for Philosophy of Science.
Soler, Léna. 2008c. “Revealing the Analytical Structure and Some Intrinsic Major Difficulties of the Contingentist/Inevitabilist Issue.” Studies in History and Philosophy of Science 39:230–41. Dordrecht: Springer, Boston Studies for Philosophy of Science.
Soler, Léna. 201X. “A General Structural Argument in Favor of the Contingency of Scientific Results.” In Science as It Could Have Been. Discussing the Contingent/Inevitable Aspects of Scientific Practices, edited by L. Soler, E. Trizio, and A. Pickering. In progress.
Van Bendegem, Jean-Paul, Amirouche Moktefi, Valéria Giardino, and Sandra Mols, eds. 2012. From Practice to Results in Logic and Mathematics. Philosophia Scientiae, Special Issue, 16(2), February 2012.
Wimsatt, William. 1981. “Robustness, Reliability and Overdetermination.” In Scientific Inquiry and the Social Sciences, edited by M.B. Brewer and B.E. Collins, 125–63. San Francisco, CA: Jossey-Bass Publishers. Reprinted in (Wimsatt 2007a), 43–71.
Wimsatt, William. 2007a. Re-engineering Philosophy for Limited Beings, Piecewise Approximations to Reality. Cambridge, MA, and London, England: Harvard University Press.
Wimsatt, William. 2007b. Robustness and Entrenchment, How the Contingent Becomes Necessary. Re-engineering Philosophy for Limited Beings, Piecewise Approximations to Reality, Chapter 7, 133–45. Cambridge, MA and London, England: Harvard University Press.
Acknowledgements
Concerning the content of this introduction, I am grateful to Jacob Stegenga and Thomas Nickles for their useful comments. Many thanks also to them, and to Emiliano Trizio, for their corrections and suggestions of improvement concerning the English language.
More generally, my personal research on robustness has benefited from a collective project, called ‘PratiScienS’, which I initiated in 2007 in Nancy, France, and have led since that time. The aim of the PratiScienS group is to evaluate what we have learned about science from the practice turn in the studies devoted to science. The issue of robustness is one of the central axes of the project. I am grateful to the members of the group for fruitful exchanges on the subject.
The PratiScienS project is supported by the ANR (Agence Nationale de la Recherche), the MSH Lorraine (Maison des Sciences de l’Homme), the Région Lorraine, the LHSP – Laboratoire d’Histoire des Sciences et de Philosophie – Archives Henri Poincaré (UMR 7117 of the CNRS) and the University of Nancy 2. The support of these institutions enabled the PratiScienS group to organize, in June 2008 in Nancy, a conference on robustness to which many contributors of the present book participated.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer Science+Business Media B.V.
About this chapter
Cite this chapter
Soler, L. (2012). Introduction: The Solidity of Scientific Achievements: Structure of the Problem, Difficulties, Philosophical Implications. In: Soler, L., Trizio, E., Nickles, T., Wimsatt, W. (eds) Characterizing the Robustness of Science. Boston Studies in the Philosophy of Science, vol 292. Springer, Dordrecht. https://doi.org/10.1007/978-94-007-2759-5_1
Download citation
DOI: https://doi.org/10.1007/978-94-007-2759-5_1
Published:
Publisher Name: Springer, Dordrecht
Print ISBN: 978-94-007-2758-8
Online ISBN: 978-94-007-2759-5
eBook Packages: Humanities, Social Sciences and LawPhilosophy and Religion (R0)