Search

espanolEl Positivismo Logico a menudo ha sido presentado como una corriente filosofica monolitica obviando las profundas disensiones que hubo en su seno. El presente articulo pretende corregir esta vision. Los dos primeros apartados de este articulo describen la constitucion del movimiento y los elementos supuestamente nucleares de la concepcion filosofica positivista. El tercer apartado discute las discrepancias sobre el papel de la filosofia entre los autores mas representativos del movimiento (Schlick, Carnap y Neurath). Para concluir se sugiere que la vigencia del legado positivista reside no tanto en alguna tesis filosofica en particular, sino mas bien en su compromiso metafilosofico con valores como la precision y la claridad. EnglishLogical positivism has been usually presented as a monolythic philosophical trend even though fundamental disagreements among its members were a matter of fact. This paper is focused precisely on those internal discrepancies. The historical setting and the allegedly fundamental philosophical claims are dealt with in the first two sections. The third section is focused on the notorious discrepancies among Schlick, Carnap and Neurath on the role assigned to philosophy. Finally, I suggest that the legacy of Logical Positivism lies not so much in any particular philosophical thesis, but rather in its metaphilosophical commitment to values such as precision and clarity.

Epidemiological models have played a central role in the COVID-19 pandemic, particularly when urgent decisions were required and available evidence was sparse. They have been used to predict the evolution of the disease and to inform policy-making. In this paper, we address two kinds of epidemiological models widely used in the pandemic, namely, compartmental models and agent-based models. After describing their essentials—some real examples are invoked—we discuss their main strengths and weaknesses. Then, on the basis of this analysis, we make a comparison between their respective merits concerning three different goals: prediction, explanation, and intervention. We argue that there are general considerations which could favour any of those sorts of models for obtaining the aforementioned goals. We conclude, however, that preference for particular models must be grounded case-by-case since additional contextual factors, as the peculiarities of the target population and the aims and expectations of policy-makers, cannot be overlooked.

Extrapolation of causal claims from study populations to other populations of interest is a problematic issue. The standard approach in experimental research, which prioritises randomized controlled trials and statistical evidence, is not devoid of difficulties. Granted that, it has been defended that evidence of mechanisms is indispensable for causal extrapolation. We argue, contrarily, that this sort of evidence is not indispensable. Nonetheless, we also think that occasionally it may be helpful. In order to clarify its relevance, we introduce a distinction between a positive and a negative role of evidence of mechanisms. Our conclusion is that the former is highly questionable, but the latter may be a trustworthy re- source for causal extrapolation.; La extrapolación de relaciones causales de poblaciones de estudio a otras poblaciones de interés es una cuestión problemática. El procedimiento estándar en investigación experimental, el cual prioriza los en- sayos controlados aleatorizados y la evidencia estadística, no está carente de dificultades. Dada esta situación, se ha planteado que la evidencia de mecanismos es indispensable para la extrapolación causal. Nosotros argumentamos que, por el contrario, este tipo de evidencia no es indispensable. Sin embargo, pensamos que puede ser de ayuda en ciertas ocasiones. Para clarificar su relevancia, distinguimos entre el rol positivo y el rol negativo de la evidencia de mecanismos. Nuestra conclusión es que el primero es altamente cuestionable, pero el segundo puede ser un recurso fiable para la extrapolación.

According to Thomas Kuhn, theory-choice is guided by a set of shared values. However, scientists differ concerning the way these values are interpreted. They also differ about the respective weight given to those values. Kuhn’s conclusion is, then, that theory-choice is an inherently subjective decision. By taking into account recent methodological practice, this essay argues that subjectivism can be minimized. Model-selection and Bayesian inference are subsequently considered. Criteria for comparing models show that values are not independently pursued while Bayesian inferential apparatus can be applied so that subjective factors are partially neutralized.

According to the “no-miracles argument” (NMA), truth is the best explanation of the predictive-instrumental success of scientific theories. A standard objection against NMA is that it is viciously circular. In Scientific Realism: How Science Tracks Truth Stathis Psillos has claimed that the circularity objection can be met when NMA is supplemented with a reliabilist approach to justification. I will try to show, however, that scientific realists cannot take much comfort from this policy: if reliabilism makes no qualifications about the domain where inference to the best explanation is reliable, scientific realists flagrantly beg the question. A qualified version of reliabilism, on the other side, does not entitle us to infer the realist conclusion. I conclude, then, that Psillos’s proposal does not make any significant progress for scientific realism.

This paper deals with inference guided by explanatory considerations –specifically with the prospects for a probabilistic interpretation of it. After pointing out some differences between two sorts of explanatory reasoning – i.e.: abduction and “inference to the best explanation” – in the first section I distinguish two tasks: (a) to discern which explanation is the best one; (b) to assess whether the best explanation deserves to be legitimately believed. In Sect. 20.2 I discuss some recent definitions of explanatory power based on “reduction of uncertainty” (Schupbach and Sprenger 2011; Crupi and Tentori 2012). Even though a probabilistic framework is a promising option here, I will argue that explanatory power so defined is not a convincing characterization of what makes a particular hypothesis better, from an explanatory point of view, that an alternative option. Then, in Sect. 20.3 I will suggest a sufficient condition (rule R1*) as my answer to (a). Regarding (b) I will propose a probabilistic threshold as a minimal condition for entitlement to believe (Sect. 20.4). The rule R1* and the threshold condition are intended as a partial explication of explanatory value (and, consequently, also as a partial explication of “inference to the best explanation”).

Van Fraassen's argument from the "bad lot" challenges realist interpretations of inference to the best explanation (IBE). In this paper I begin by discussing the replies suggested by S. Psillos and P. Lipton. I do not find them convincing. However, I think that van Fraassen's argument is flawed. First of all, it is a non sequitur. Secondly, I think that the real target for the scientific realist is the underlying assumption that epistemic justification results from a comparative assessment among rival explanations. I argue that justification for believing an explanation does not depend on comparison, but on the extent that criteria of explanatory goodness are fulfilled. Therefore, in addition to offering more or less intuitive IBE-tailored arguments, realists fond of IBE should have to analyze the implicit standards of explanatory goodness. In the last section I distinguish between contextual and transcontextual criteria concerning explanatory goodness. Concerning the latter, I focus on consilience, simplicity, analogy and conservatism.

Bas van Fraassen’s recent endorsement of empiricist structuralism is based on a particular approach to representation. He sharply distinguishes between what makes a scientific model M a successful representation of its target T from what makes M a representation of T and not of some other different target T’. van Fraassen maintains that embedment (i.e.: a particular sort of isomorphism which relates structures) gives the answer to the first question while the user’s decision to employ model M to represent T accounts for the representational link. After discussing the rationale for this approach, I defend that indexical constraints like those favoured by van Fraassen cannot be the last word concerning what makes a scientific model a representation of something in particular. Rather, I argue that (i) the representational role of models — at least of scientific models — is inextricably related to their ability to convey some knowledge about their purported target, and (ii) this is an effective constraint on the user’s decisions. Both claims cast some doubt on the aforementioned distinction insofar as not only success in representation, but also the existence of a representational relation, is rooted in our knowledge about the target.

Philip Kitcher has developed a sort of inductivist-reliabilist justification for scientific realism. After distinguishing his argument from a wellknown abductivist one (the “no-miracles” argument), I will argue that Kitcher’s proposal cannot adequately meet the antirealist challenge. Firstly, it begs the question against the antirealists; secondly, it can hardly support a plausible – piecemeal – scientific realism. I will explore an alternative inductivist approach that exploits correlations between theoretical properties and empirical success. On my view, its prospects for avoiding the aforementioned shortcomings are better than Kitcher’s standpoint. I dare say, however, that an inductivist strategy alone cannot satisfy the demands of scientific realism since, in the end, an abductive move may well be mandatory for grounding it.

The paper analyzes which is the proper role of history of science in philosophy of science. I argue that, traditionally, philosophy of science has focused on the normative dimension of science. Several positions within philosophy of science stem from different perspectives on such dimension. I classify them according to two variables (descriptivism/prescriptivism, and historicism/non-historicism). Then I discuss the role of history of science within all these four alternatives. I conclude that: (1) philosophers and historians pursue not only different, but opposite aims; (2) historical evidence is useful for assessing the relation between methods and results; besides this, only very simple or radical philosophical views can be ¿refuted¿ by historical evidence; (3) the disagreement between those who think that history of science is necessary for philosophy of science and those who think it is not is, in fact, a disagreement about the prospects for a general theory of science; (4) such disagreement cannot be resolved appealing to historical evidence; if it can be resolved, philosophical arguments should somehow be invoked.

El articulo analiza dos enfoques coherentistas de lajíístifícación (K. Lehrer y L. Boniour). Ambas versiones se consideran insatisfactorias. En primer lugar, sobrecargan el aparato introspectivo del sujeto cpistémi co, lo cual impide dar cuenta de nuestras intuiciones preanalíticas sobre sujetos "limitados" (niños o animales>. Por otro lado, no consiguen esta blecer un viculo convincente entre justificación ----entendida como cohe í'encia--- y verdad entendida como correspondencia.

RESUMENUna forma común de entender el realismo científico (RC) en las últimas décadas ha sido plantearlo como una inferencia explicativa: RC es la mejor explicación del éxito predictivo-instrumental de la ciencia. Algunos de sus partidarios mantienen, además que es una hipótesis empíricamente constrastable. Intentaré argumentar, que, entendido así, RC no es empíricamente contrastable. En primer lugar, aunque el éxito predictivo-instrumental initerrumpido de una teoría T es una consecuencia observacional de la verdad de T, este hecho no hecho no constituye una evidencia empírica diferente del propio "explanandum". En segundo lugar, elaorar un registro histórico del éxito -no sólo empírico, sino teórico- obtenido mediante la postulación de entidades por consideraciones explicativas, confirmaría como mucho, y eso suponiendo que fuera posible, una cocincidencia entre una metodología determinada y unos resultados, pero no daría cuenta del vínculo explicativo entre éxito predictivo instrumental por un lado, y verdad y existencia, por otro. Por consiguiente, RC no es una hipótesis empírica en un sentido genuino; a fortiori, tampoco es una hipótesis científica. Esta conclusión, no obstante, no cierra el camino a un realismo científico de carácter local.PALABRAS CLAVEREALISMO CIENTÍFICO, OBSERVACIÓN, TEORÍA, INFERENCIA A LA MEJOR EXPLICACIÓNABSTRACTA common way of understanding scientific realism (SR) during the latest decades says that SR is the best explanation of the predictive success enjoyed by scientific theories. Some os this advocates claim, aslo, that SR is an empirically testable hypothesis. I will try to argue that, as an explanation of predictive sucess, SR is not empirically testable. Firstly, even though the uninterrupted preditive success of T is an observational consequence of T´s truth, this fact is not a kind of evidence distinguishable from the very explanandum. Secondly, a historical record of success obtained by postulating theoretical emities would confirm, at most, a correlation between some methodological norms and some particular results. But confirming such correlation is not the same as vindicating an explanotory link between truth and existence (the explanans), and predictive success (the explanandum). In sum, SR is not a genuine empirical hypothesis; a fortiori, it is neither a scientific hypothesis. Anyway, this conclusion does not forbid some kind of "local" scientific realism.KEYWORDSSCIENTIFIC REALISM, OBSERVATION, THEORY, INFERENCE TO THE BEST EXPLANATION

El artículo consta de cuatro apartados. El primero hace un breve recorrido histórico para mostrar la estrecha imbricación entre los juegos de azar y las matemáticas. En el segundo se argumenta que los juegos de azar son 'injustos' para el apostante y favorables para la casa de apuestas; precisamente en eso reside el margen de benefi cios que obtiene esta última y que la convierte en un negocio rentable. Se explica además cómo, mediante un sencillo cálculo, podemos averiguar si una apuesta en particular, o un juego, entendiendo éste como una secuencia de apuestas, es 'justo' o no. También se razona por qué cuando los apostantes se enfrentan a un 'cociente de apuestas' adverso a la larga acabarán perdiendo todo su dinero. Si el jugador advirtiera la certidumbre de su ruina a largo plazo, por razones matemáticas, quizá dejase de apostar. Pero lo que a menudo suele ocurrir es bien distinto: el jugador hace sus cábalas sobre cómo funciona el azar y sobre sus posibilidades de controlarlo, extrae sus conclusiones sobre el comportamiento del dispositivo físico que genera los resultados, y' sigue apostando. La evidencia experimental apunta además que cuando razonamos sobre probabilidades, jugadores y no jugadores, somos muy propensos a cometer errores. Por eso en el tercer apartado se exponen diversos razonamientos engañosos ''falacias'', relacionados con la probabilidad y el azar. Para acabar se incluye una breve sección con las conclusiones. The article consists of four sections. The fi rst is a short historical look to display the close relation between games of chance and mathematics. The second argues that games are chance are 'unfair' for the gambler and favourable for the betting fi rm; this is precisely where the profi t margin lies for the fi rm, and why this is a profi table business. It also explains how, by means of a simple calculation, we can fi nd out whether a particular bet or a game, understood as a series of bets, is 'fair' or not. It furthermore gives the reasons why gamblers will in the long term end up losing their money when they have to play against an adverse 'betting quotient'. If players could realise the inevitability of their ruin in the long term for mathematical reasons, they might stop betting. What often tends to happen is nevertheless quite another matter: the players work out their calculations as to how chance works and their possibilities of controlling this, draw their conclusions as to how the physical device generating the results works and ' go on betting. The experimental evidence also indicates that when gamblers and non-gamblers alike reason on probabilities, we are very prone to make mistakes. For this reason the third section puts forward several errors of reasoning ' 'fallacies', connected with probability and chance. To end with there is a short section containing the conclusions.