Une science radicalement différente est‐elle plausible ? Dépendance vis‐à‐vis du chemin suivi et contingence des résultats scientifiques

Theodore ARABATZIS (Department of History and Philosophy of Science, National and Kapodistrian University of Athens)

The Contingentism / Inevitabilism Debate and its Implications for the Historiography of Science

In this talk, in line with my earlier plea for a philosophically informed historiography of science (Arabatzis 2017), I will discuss the historiographical relevance of the philosophical debate on the contingency/inevitability of the process and products of science. My starting point will be a recent article where I attempted to bring out the many faces of understanding science historically (Arabatzis 2019). Drawing on that article, I will outline some problems of causal explanation in history of science and I will argue that its prospects are rather dim. I will then suggest an explanatory approach to past science based on the intellectual and material resources that were available to historical actors and on the constraints that delimited their available options. Those resources and constraints merely delineate a space of possibilities, within which the trajectories followed by historical actors can be explained on the basis of contingent factors, namely their biographical peculiarities and particular projects of inquiry. I will illustrate this point by examples from the history of the physical sciences.

Arabatzis, T. (2017), “What’s in it for the historian of science? Reflections on the value of philosophy of science for history of science,” International Studies in the Philosophy of Science, 31:1, 69‐82.

Arabatzis, T. (2019), “Explaining science historically,” Isis 110:2, 354‐359.

Agnes BOLINSKA (University of Cambridge)

Negotiating History: Contingency, Canonicity and Case Studies

Objections to the use of historical case studies for philosophical ends fall into two categories. Methodological objections claim that historical accounts and their uses by philosophers are subject to various biases. We argue that these challenges are not special; they also apply to other epistemic practices. Metaphysical objections, on the other hand, claim that historical case studies are intrinsically unsuited to serve as evidence for philosophical claims, even when carefully constructed and used, and so constitute a distinct class of challenge. Each of these can be parsed as a critique based on the path dependency of science. We show that attention to what makes for a canonical case can address these problems. A case study is canonical with respect to a particular philosophical aim when the features relevant to that aim provide a reasonably complete causal account of the results of the historical process under investigation—that is, when it is path dependent in the right kind of way. We show how to establish canonicity by evaluating relevant path dependencies in two prominent examples from the history of science: Eddington’s confirmation of Einstein’s theory of general relativity using his data from the 1919 eclipse and Watson and Crick’s determination of the structure of DNA.

Julien BORGEON (Archives Henri Poincaré ‐ Philosophie et Recherches sur les Sciences et les Technologies, Université de Lorraine, Nancy)

Path Dependence in Technological and Scientific Development: A Critical Comparison

To what extent can or should we say that our current scientific knowledge is contingent? Is the path taken by science, in the course of its actual history, able to have determinant effects on scientific theories and other scientific results? Or was something like the present state of our scientific knowledge inevitable, whatever the accidental details of the scientific path actually followed? The concept of “path dependence”, inherited from economics and philosophy of economics, and classically applied to technologies, can no doubt help us to progress in these questions about scientific knowledge. “Technical interrelatedness”, “economies of scale”, “quasi‐ irreversibility of investments” are among the typical historical processes theorized by economists to explain that technologies can become “locked” onto a certain path (David 1985, 334). Such “locked‐in effects” have been perceived as worrying, since they can occur even if one is aware of the inferiority, in terms of efficiency, of the locked technology compared to alternative technologies, in principle available but in practice not adopted.

I will start from an article published by Mark S. Peacock in 2009, “Path Dependence in the Production of Scientific Knowledge”, which offers one of the few existing attempts to compare path dependence in technological development and path dependence in scientific development. More precisely, in this article, Peacock examines to what extent the framework first elaborated by economists in order to characterize path dependence in technological development, often in the hope to avoid the contingent domination of inferior technological products, can be applied as well to scientific development and scientific products.

The first part of my talk will present Peacock’s analyses and main conclusions. In the second part, I will propose a critical discussion, not only of Peacock’s position, but also, more generally, of attempts to apply to the scientific domain the path dependence framework initially conceived in relation to the technological field.

David, Paul. A. 1985. “Clio and the economics of QWERTY”, American Economic Review, 75 :2, 332‐ 337.

Jeroen BOUTERSE (Leiden)

History of Science If Science Is Contingent

Does it matter, to our historical accounts of science, if we believe that the outcomes of the history we try to understand could have been significantly different? The accusation of a bias towards present‐day knowledge is usually associated with “Whig history”, which has inevitabilist connotations; but contingentism comes with its own problems. In particular, do our historical explanations risk being viciously circular, if they rely upon a body of knowledge that we believe to be contingent and whose historical development we are trying to explain?

I believe that historians can in principle embrace contingentism about science (of which I will try to formulate a definition geared towards the kind of questions within the purview of historians), while retaining a liberal attitude towards employing present‐day beliefs about the world. They need these beliefs, I will argue, for the very reason that science is about the world.

Katherina KINZEL (Temple University)

Tracking Truth from Arbitrary Starts? Path Dependence and Scientific Realism

This paper explores the relation between contingentism and scientific realism through the lens of path dependence arguments. The concept of path dependence provides a technical explication of the claim that scientific results are tied to their histories, and that these histories are contingent. It is often taken to support to the contingentist view that the history of science could have led to the emergence and acceptance of alternative yet equally successful scientific results. However, scientific realists too have argued that the historical development of science is path dependent, and on most readings, scientific realism is incompatible with contingentism. This paper seeks to gain a clearer understanding of what form of path dependence the scientific realist can embrace, and what this tells us about the nature of the disagreement between realists and contingentists.

Joseph D. MARTIN (Durham University)

What Is a Canonical Case?

In earlier work, we argue that historical cases must be rendered canonical before they can plausibly serve as evidence for philosophical claims, where canonicity is established through a process of negotiation among historians and philosophers of science to show that it is path dependent in the right kind of way. Here, we extend this proposal by suggesting how that negotiation might take place in practice. We begin with a review of existing canonicity practices in history and philosophy of science. The way in which the working stock of historical examples used to support philosophical claims is established has long been informal. As a result, it is somewhat haphazard. HPS itself, that is, is path dependent. Cases often become standard examples for reasons other than their appropriateness for the purposes at hand. We then show how the lack of rigor around canonicity has muddied the waters in selected philosophical debates. Establishing the confusion encouraged by haphazard canonicity practices then sets the groundwork for proposing ways in which they can be improved.

Gregory RADICK (University of Leeds)

How Can We Know What Might Have Been in the Scientific Past? Reflections on the Debate over Mendelism

We make judgments all the time about all kinds of claims about the might‐have‐been or counterfactual past. At least some of the time, those judgments take into account evidence from the actual course of history. Concentrating on claims about the counterfactual scientific past, this talk will consider the kinds of evidence that can bear in deciding questions about whether a given scientific achievement was inevitable, i.e., independent of its particular history, or whether, had that history gone differently, science might have gone differently too. The kinds of evidence to be considered are: (i) evidence from histories converging on the scientific achievement in question; (ii) evidence from histories diverging from that achievement; (iii) evidence of alternatives under contemplation at the time; and (iv) evidence from attempts to develop those alternatives in the present. Throughout I’ll draw on examples from my own studies of the post‐1900 debate over Mendel, including, under (iv), an experiment assessing the effects of teaching genetics from a Mendelian or a “Weldonian” starting point. I’ll suggest that the availability of (iv) potentially puts counterfactual histories of science on a more secure epistemic footing than other kinds of counterfactual history. And I’ll stress that, beyond (i)‐(iv), historians of science pursuing a counterfactual thesis can avail themselves of all the evidence relating to what was dependent on what in the post‐achievement history of science.

Léna SOLER (Archives Henri Poincaré ‐ Philosophie et Recherches sur les Sciences et les Technologies, Université de Lorraine, Nancy)

We Could Be Committed to an Alternative Incompatible Physics – Historically Plausible, Epistemologically Legitimate

The talk will back up the following contingentist thesis: one of our current most fundamental physical paradigm, namely, the current paradigm of quantum physics, including what is valued today as “the” standard quantum theory, is the contingent product of an historical path that could have been significantly different. The latter phrase, “the path could have been significantly different”, must be understood as involving, not only an empirical possibility (i.e., some significantly different paths  were historically plausible), but also an epistemologically legitimate possibility (i.e., at least some of the alternative historically plausible scientific paths and the emerging incompatible physical results are not reducible to wrong tracks, theoretical mistakes, methodological faults or anything similar). Had such alternative paths been undertaken instead of the one actually followed, we might live today with an alternative microscopic physics, deeply different, at several important levels, from the microscopic physics currently presented as “the” right one. In this vein, according to some historically plausible and epistemologically legitimate counterfactual scenarios, we would take for granted an alternative physical theory literally incompatible with “the” currently standard quantum theory. More precisely, assuming a literal reading of scientific theories (as most scientists are usually inclined to assume), we would believe in descriptions of the physical reality in direct contradiction with what most physicists, most people and many philosophers of science believe, today, about the “furniture” and behaviour of the physical world relying on ‘the” standard quantum mechanics. Moreover, practicing physicists would deal with significantly different standard problems and standard solutions, and would have to cope with significantly different key difficulties. Last but not least – and still more consequential regarding the contingency issue as we shall see – most physicists would be committed to a conception of physics, and more generally to “norms of scientificity”, in conflict with the ones officially and more or less explicitly promoted nowadays, in particular in scientific curricula. The conflict would affect commitments and judgments, simultaneously and interdependently, about what is a satisfying physical theory, about what merits the name of physical knowledge, and about what has the right to count as physical reality.

To defend this contingentist thesis, I will rely, inspired by James Cushing’s work, on the counterfactual history of quantum physics. Several variants of contrary‐to‐facts scientific trajectories will be introduced, each bringing about a similar alternative physics incompatible with the currently standard quantum physics, and arguments will be provided to warrant both the historical plausibility and the epistemological legitimacy of the corresponding virtual scenarios. Reflecting on such might‐have‐ been scientific scenarios, and considering some factors decisively involved not only in them but, more generally, in the history of science as we know it, and more especially in scientific education as we know it, I will show that a certain sort of path dependence cannot but be decisively involved in scientific development – we could talk, here, of something like a “quasi‐inevitability” of a historically‐relativized transcendental type, that is, a quasi‐inevitability relative to some “definitory” conditions under which, as a matter of facts, science is practiced and conceived in our world. This sort of path dependence, I will then argue, supports a strong and philosophically consequential contingentist thesis about scientific achievements. The thesis convincingly applies to the particular case of quantum physics under scrutiny, but could potentially be generalized – admittedly, under further discussions. Finally, insights will be provided concerning the issue of whether we should worry about the corresponding path dependence of scientific processes and the related contingency of scientific products, and some possible epistemological and political incidences will be sketched.

Luca TAMBOLO (Bologna)

Background knowledge, multiple discoveries, inevitability

Background knowledge (the set of previously acquired beliefs that, at any point, a scientific community has to take as unproblematic within the investigation of a certain problem) is one way in which path dependence (broadly, the weight of the past on current practice) manifests itself in scientific inquiry. In this talk, after emphasizing the role that background knowledge plays in bringing about cases of multiple discoveries, I investigate two questions concerning the value of results of scientific inquiry that are multiply arrived at. The first is whether one should maintain that multiple discoveries have something epistemically special, setting them apart from results arrived at only once. The second is whether multiple discoveries can be viewed as instantiations of the inevitability thesis, according to which the successful investigation of a domain yields results that, being correct, will feature in end‐run science. I answer the first question in the negative, arguing that since any result of scientific inquiry must be assessed based on appropriate standards, which do not change depending on whether the result is independently arrived at by multiple researchers or by just one, nothing sets multiple discoveries apart. Concerning the second question, I analogously argue that since any result of scientific research that qualifies as correct in the relevant sense will feature in the end‐run science to which the inevitability thesis refers, no matter how many times the result has been independently reached, nothing sets multiple discoveries apart. Therefore, I suggest, the temptation to view multiple discoveries as a phenomenon favoring the inevitability thesis over the contingency thesis—a temptation that the scientific realist may well have—should be resisted to.