In recent years there has been a revival of the debate about the relation between history and philosophy of science. This article seeks to contribute to the discussion by approaching the issue from a new angle. To rethink the relation between the two domains of study, I apply an important insight about scientific practice to the practice of integrating the history and philosophy of science: the insight that the scientific paper does not give a faithful account of the actual research (...) pursued in the laboratory or in the field. Arguably, the scholarly article about science is also not a transparent window to the activity of producing such an article. But if it is not, we need to redirect our attention. If we want to understand the nature and merits of integrating the history and philosophy of science, we need to examine both the actual activity of integrating and the scholarly paper produced by it. To consider what one can learn from such an inquiry, I reflect on my own activity of studying scientific justification through the combination of historical and philosophical analysis. Probing a concrete historical episode, micro-anatomical research on the retina in the 19th century, I pursue two related questions, the first-order question, 'What exactly is scientific justification in the given case?' and the meta-question, 'How do I go about analysing justificatory practices?' I then characterise the nature of my analysis and consider what can be learned from the study of the practice of integrating the history and philosophy of science. (shrink)

Allan Franklin has identified a number of strategies that scientists use to build confidence in experimental results. This paper shows that Franklin's strategies have direct analogues in the context of computer simulation and then suggests that one of his strategies—the so-called 'Sherlock Holmes' strategy—deserves a privileged place within the epistemologies of experiment and simulation. In particular, it is argued that while the successful application of even several of Franklin's other strategies (or their analogues in simulation) may not be sufficient for (...) justified belief in results, the successful application of a slightly elaborated version of the Sherlock Holmes strategy is sufficient. (shrink)

With the careful use of the polywater episode in the history of chemistry as a case study, I will show that the distinction recently made in the philosophy of science between experimenting on an entity and manipulating that entity is best seen as a distinction between experimenting on, and experimenting with, that entity. The polywater case also reveals that Ian Hacking's 1983 manipulability criterion is not a necessary condition for realism, and that scientists can, and do, justifiably change their minds (...) as to the reality of theoretical entities. (shrink)

I present a justification for the intution that more-varied data are more valuable than the same number of less-varied data by showing that the more-varied data help to improve the accuracy of our predictions.

This paper is concerned with the problem of experimental error. The prevalent view that experimental errors can be dismissed as a tiresome but trivial blemish on the method of experimentation is criticized. It is stressed that the occurrence of errors in experiments constitutes a permanent feature of the attempt to test theories in the physical world, and this feature deserves proper attention. It is suggested that a classification of types of experimental error may be useful as a heuristic device in (...) studying the nature of these errors. However, the standard classification of systematic and random errors is mathematically based does not focus on the causes of the errors, their origins, or the contexts in which they arise. A new typology of experimental errors is therefore proposed whose criterion is epistemological. This typology reflects the various stages that can be discerned in the execution of an experiment, each stage constituting a category of a certain type of experimental error. The proposed classification consists of four categories which are illustrated by historical cases. (shrink)

On the basis of an analysis of a single paper on plate tectonics, a paper whose actual content is nowhere in evidence, Frederick Suppe concludes that no standard model of confirmation—hypothetico-deductive, Bayesian-inductive, or inference to the best explanation—can account for the structure of a scientific paper that reports an experimental result. He further argues on the basis of a survey of scientific papers, a survey whose data and results are also absent, that papers which have a rather stringent length limit, (...) such as the one on plate tectonics, are typical of science. Thus, he concludes that no standard confirmation scheme is capable of dealing with scientific practice. Suppe also requires that an adequate model of philosophical testing should be able to account for everything in such scientific papers, in which space is at a premium. (shrink)

Several different Bayesian models of epistemic utilities (see, e. g., [37], [24], [40], [46]) have been used to explain why it is rational for scientists to perform experiments. In this paper, I argue that a model-suggested independently by Patrick Maher [40] and Graham Oddie [46]-that assigns epistemic utility to degrees of belief in hypotheses provides the most comprehensive explanation. This is because this proper scoring rule (PSR) model captures a wider range of scientifically acceptable attitudes toward epistemic risk than the (...) other Bayesian models that have been proposed. I also argue, however, that even the PSR model places unreasonably tight restrictions on a scientist's attitude toward epistemic risk. As a result, such Bayesian models of epistemic utilities fail as normative accounts-not just as descriptive accounts (see, e. g., [31], [14])-of scientific inquiry. (shrink)

We appeal to the theory of Bayesian Networks to model different strategies for obtaining confirmation for a hypothesis from experimental test results provided by less than fully reliable instruments. In particular, we consider (i) repeated measurements of a single test consequence of the hypothesis, (ii) measurements of multiple test consequences of the hypothesis, (iii) theoretical support for the reliability of the instrument, and (iv) calibration procedures. We evaluate these strategies on their relative merits under idealized conditions and show some surprising (...) repercussions on the variety-of-evidence thesis and the Duhem-Quine thesis. (shrink)

Glenn Harrison [Journal of Economic Methodology, 2013, 20, 103–117] discusses four related forms of methodological intolerance with respect to field experiments: field experiments should rely on some form of randomization, should be disconnected from theory, the concept of causality should only be defined in terms of observables, and the role of laboratory experiments is dismissed. As is often the case, the cause of intolerance is ignorance, as it is here. To acquire knowledge about potential influences, which we need for both (...) the evaluation of internal and external validity of experimental results, we cannot do without theory. A purely empiricist methodology will be unable to give us sufficient understanding of the validity of these results. An account of causality only based on directly observed things, is an account based on factual influences only. This account will be too restricted, because it will not deal with the unobserved potential influences, which we need – again – for the evaluat... (shrink)

Laboratory experimentation was once considered impossible or irrelevant in economics. Recently, however, economic science has gone through a real ‘laboratory revolution’, and experimental economics is now a most lively subfield of the discipline. The methodological advantages and disadvantages of controlled experimentation constitute the main subject of this thesis. After a survey of the literature on experiments in philosophy and economics, the problem of testing normative theories of rationality is tackled. This philosophical issue was at the centre of a famous controversy (...) in decision theory, during which a methodology of normative falsification was first articulated and used to assess experimental results. In the third chapter, the methodological advantages of controlled experimentation are illustrated and discussed with examples taken from the experiments on the so-called ‘preference reversal’ phenomenon. Laboratory testing allows to establish with a high degree of certainty that certain phenomena lie behind the experimental data, by means of independent testing, elimination of alternative hypotheses, and the use of different instruments of observation. The fourth chapter is devoted to a conceptual analysis of the problem of ‘parallelism’. This is the problem of inferring from the occurrence of a phenomenon in the laboratory, to its instantiation also in non-laboratory environments. Experimental economists have discussed parallelism at length, and their views are presented and criticised. Eventually, it is argued that parallelism is a factual matter and as such can only be established on empirical grounds. The fifth chapter provides an example of how one can argue for parallelism, focusing on the case of experimentation on the ‘winner’s curse’ phenomenon. The role of experiments as ‘mediators’ between theoretical models and their target domain of application is illustrated, and the structure of parallelism arguments analysed in detail. Finally, in the last chapter, economic experiments are compared to simulations, in order to highlight their specific characteristics. (shrink)