Mostrando entradas con la etiqueta 2010. Mostrar todas las entradas
Mostrando entradas con la etiqueta 2010. Mostrar todas las entradas

18/1/11

Jean-Claude Passeron, Le raisonnement sociologique. Un espace non-poppérien de l’argumentation, París, Albin Michel, 2006.

En el prefacio a la segunda edición francesa, Jean-Claude Passeron nos advierte de los múltiples malentendidos que lastraron el debate en torno a Le raisonnement sociologique (LRS ). Creo que la densidad conceptual de la obra explica, si no justifica, muchos de ellos, al menos en mi caso. Pese a las aclaraciones añadidas a esta nueva edición, me temo que sólo puedo contribuir a este debate aportando nuevos malentendidos que le den al autor la oportunidad de elucidarlos. La novedad de estos malentendidos, si es que hay alguna, radica en la diferencia de perspectivas entre Passeron y el autor de estas líneas, muy probablemente generacional. Yo comenzaba mis estudios universitarios cuando se publicaba la primera edición de LRS y leo ahora la segunda después de tan solo una década dedicado a la filosofía de las ciencias sociales. De ahí mi sorpresa no ya ante las tesis metodológicas de LRS, sino ante la justificación que Passeron nos propone.

Una de sus tesis principales, según la (mal)interpreto, es que las ciencias sociales tienen que servirse necesariamente de argumentos informales, pues es imposible aislar de modo unívoco y dar una definición general todas las variables pertinentes para analizar matemáticamente una situación eminentemente singular. Como justificación, Passeron apela a la inviabilidad del ideal científico defendido originalmente por el Círculo de Viena, de un lado, y por Popper, de otro. Como es sabido, este ideal se basaba en una concepción formal de las teorías que se demostró indefendible, por razones que Passeron desarrolla con amplitud en un epílogo que recapitula su propia posición en LRS. Y de ahí mi sorpresa, y quizá el primer malentendido: ¿quién sostenía en 1991 las tesis que Passeron critica?

Me temo que se trata de una querella de sociólogos, más que un debate estrictamente filosófico. A la altura de 1960, autores como Carl Hempel o Ernst Nagel sabían ya de las dificultades de justificar la superioridad del conocimiento científico (frente a la metafísica) a partir de la estructura de sus teorías y ensayaron una nueva vía que es, aparentemente, la que aquí quiere seguir Passeron: analizar en qué condiciones resultan aceptables los distintos tipos de explicación científica, concebidos como otras tantas formas de argumentación. Es decir, pasos inferenciales, no siempre deductivos desde un conjunto de premisas a una conclusión. Durante los últimos 40 años, la filosofía de las ciencias sociales se sirvió ampliamente de esta estrategia generando un cuerpo de debates sobre la potencia argumental de las explicaciones que nos vienen ofreciendo economistas, sociólogos, antropólogos, etc. Y esto es lo que un lector de mi generación/educación habría esperado encontrar en LRS: no tanto la crítica del proyecto positivista original, como una tipología de los argumentos que, según Passeron, caracterizarían el razonamiento sociológico, junto con una discusión de su fortaleza .

Pero se diría que a Passeron le interesa más bien mostrar, a través de sus críticas al formalismo logicista del positivismo, el carácter necesariamente incompleto del formalismo matematizante en ciencias sociales. Y la fuerza de su propio argumento se apoya en las dificultades semánticas de semejantes proyectos: e.g., la imposibilidad de construir un “vocabulario observacional” en el que volcar sin ambigüedad los datos que arroje la investigación empírica, de modo que su acumulación sirva como base para contrastar teorías sociológicas o construir generalizaciones legiformes. Este sería mi segundo malentendido: ¿tienen alguna vigencia estos argumentos o se reeditan, como indica el autor (p. 22), simplemente para documentar la Historia de los debates metodológicos en Francia? En una época en la que la explotación sistemática de bases de datos y los experimentos sobre decisiones individuales son ya objeto de conversación popular gracias a éxitos de venta como Freakonomics o Predictably irrational, ¿cabe sostener todavía las posiciones de LRS tal como se formularon en 1991? Los más críticos con semejantes empresas son justamente los teóricos más formalistas en las ciencias sociales (los economistas), pues ponen de manifiesto cómo con un aparato teórico mínimo es posible extraer conclusiones interesantes a partir de datos estadísticos ajenos a la propia teoría. Por usar el famoso ejemplo de Levitt, los patrones de respuesta observados en los miles de cuestionarios realizados en las escuelas de Chicago permiten conjeturar qué profesores hacen trampa y rectifican los exámenes de sus alumnos para evitar ser penalizados por sus bajos resultados. ¿Por qué no habríamos de aceptar el contenido de esta base de datos como un vocabulario observacional de uso común en ciencias sociales?

La respuesta no está, creo, en el Círculo de Viena o en sus más inmediatos epígonos, sino en la tradición hoy más viva en filosofía de la ciencia, cuyos orígenes se remontan nuevamente a la década de 1960. Fue entonces cuando autores como Patrick Suppes se preguntaron si las dificultades que plantea el problema de la carga teórica de la observación (extensamente discutido en LRS) no se atenuarían si se recurre al álgebra, antes que a la lógica, para analizar las teorías científicas. Con ello se abandonaría, por un lado, la perspectiva lingüística que dominó la tradición positivista y, por otro, se podría tratar con mayor fidelidad la práctica científica en la que predomina el uso de modelos. Suppes llamó la atención sobre la existencia de modelos centrados exclusivamente en el procesamiento de datos empíricos (e.g., estadísticos) y, por tanto, independientes de las teorías que se aplican sobre ellos. Es decir, no absolutamente independientes respecto de cualquier teoría, pero sí respecto del aparato conceptual que se ha de aplicar sobre tales modelos de datos. Su intuición fue ampliamente desarrollada tanto en la escuela de Stanford (Cartwright, Hacking, etc.) como, formalmente, por el enfoque estructuralista (Sneed, Moulines, etc.). Así, en el caso de las bases de datos utilizadas por Levitt no pueden presumirse sesgos de la teoría económica en su generación (aunque haya otros) y en esa medida es interesante su análisis económico, por minimalista que sea el aparato teórico del autor.

Una de las principales virtudes de estos modelos de datos es la de exhibir regularidades fenomenológicas que aparecen en los datos obtenidos a partir de experimentos y otros estudios empíricos. Puede que no contemos todavía con teorías generales para dar cuenta de tales regularidades y su alcance es, desde luego, contextual. Pero su sola existencia permite el tipo de debates metodológicos que LRS parece declarar impracticables :
La vulnerabilidad y, por tanto, la pertinencia empíricas de los enunciados sociológicos sólo pueden ser definidas en una situación de extracción de información sobre el mundo que es la de la observación histórica, nunca la de la experimentación (LRS, p. 554, traducción de J. L. Moreno Pestaña).

Una réplica inmediata a mi objeción es que me este tipo de regularidades quizá existan en otros dominios de las ciencias sociales, pero no en sociología. Como antes apuntaba, Passeron defiende la historicidad del análisis sociológico de un modo tal que parece no haber lugar para aislar regularidades en los datos agregados o las decisiones individuales. Creo que esta posición se deriva, en buena parte, de una actitud anti-naturalista muy arraigada en sociología (e.g., p. 81), para la cual la universalidad que podemos encontrar en ciertos patrones de decisión de un agente no sería objeto propio de la disciplina. Pero la oleada naturalista sobre las ciencias sociales provocada por el desarrollo de la etología y la neurología durante las dos últimas décadas está poniendo de manifiesto que dentro de la Historia hay espacio para explicaciones que parten directamente de nuestra constitución biológica. Por ejemplo, nuestra miopía para estimar en qué medida se renuevan los recursos ecológicos de los que dependen nuestras sociedades, documentada sistemáticamente a lo largo de los siglos en los casos reunidos por Jared Diamond en Colapso . Los efectos sociales de este déficit cognitivo han sido ampliamente discutidos por los historiadores, pero sólo cuando incorporamos una perspectiva evolucionista sobre nuestra psicología podemos entender con precisión el mecanismo generador de esta miopía ―en lugar de atribuírselo a nuestra irracionalidad, rapacidad, etc. Cuál sea su alcance de este tipo de análisis para la Historia está todavía en discusión, pero su impacto parece suficiente como para reconsiderar si la historicidad debe cifrarse tan sólo en la ausencia de repeticiones espontáneas o en la imposibilidad de aislar las variables relevantes en un laboratorio .

No sé si acierto en mi lectura de LRS, pero no estoy en desacuerdo con las tesis de Passeron: las ciencias sociales se han de servir necesariamente de argumentos informales, cuyo alcance depende, generalmente, del contexto y su aplicación empírica está condicionada por la dificultad de controlar los factores causales que controlan los acontecimientos analizados. El problema es que, así enunciadas, no se me ocurren hoy muchos partidarios de las tesis contrarias. Y, por otro lado, los argumentos de los que se sirve para justificarlas me resultan menos convincentes que las alternativas que vengo enumerando. Los argumentos importan, pues señalan el auténtico alcance del desacuerdo: si actualizamos las referencias de Passeron para incluir el enfoque semántico en filosofía de la ciencia y limamos su anti-naturalismo, tendríamos un espacio argumental “anti-popperiano” en el que cabe una sociología que se apoyase en regularidades empíricas construidas a partir de análisis estadísticos y experimentos para construir explicaciones apelando, entre otros, a mecanismos biológicos propios de toda la especie. No es precisamente la que Passeron practica y defiende en LRS, ni tampoco pretendo yo ahora defender tal alternativa sociológica. Simplemente creo que sus argumentos no son lo suficientemente poderosos para excluir semejante alternativa y, me temo, que si uno concede más peso al debate metodológico actual que a Windelband y el Círculo de Viena no queda más remedio que tomarla en consideración. Otra cosa es que a los sociólogos les interese, pero eso no me corresponde a mí juzgarlo.

[Debate en la RES a propósito de la versión castellana de J. Moreno Pestaña, de próxima aparición en Siglo XXI con F. Aguiar y F. Vázquez y respuesta del propio Passeron]

Nancy J. Nersessian, Creating Scientific Concepts, Cambridge, MA: MIT Press, 2008.

For more than two decades now Nancy Nersessian has been working at the intersection of philosophy of science and the cognitive sciences, investigating how scientists actually think at work. Creating Scientific Concepts provides a general introduction to her approach, covering, on the one hand, historical research on the ways of thinking of 19th century physicists (namely, James Clerk Maxwell) and, on the other hand, an analysis of the cognitive foundations of scientific modelling. Nersessian’s essay is more than introduction though. Her goal is to explain conceptual innovation in science. More precisely, to articulate an analytic framework to account for the “specific modelling practices that historical records implicate in problem solving leading to conceptual innovation, specifically, analogical modelling, visual modelling and thought-experimenting” (p. 13). Let me spell this out.

Nersessian adopts an empiricist view of concepts, drawing mainly on the work of Lawrence Barsalou (Chapter 4). According to this latter, concepts would be perceptual symbols, neural correlates of sensorimotor experience (p. 124). These symbols constitute analogical representations and the sensioriomotor processes in which they originate are re-enacted whenever we use them to think. In our concepts about physical systems, the analogy captures the constraints we discern in the phenomena (e.g., causal structures). These constraints are then preserved in our mental simulations, even if we change other properties and relations of our physical concepts in order to solve whatever problem we are dealing with. The possession of a concept implies thus the skill “for constructing a potentially infinite number of simulations” according to our needs and goals (p. 126). Nersessian also argues for a coupling between external and internal representations: mental simulation often needs real-world resources “outside” our heads. However, she acknowledges that the nature of the cognitive mechanisms at the interface of this coupling is still to be articulated. Images ―diagrams, for instance― provide representational tools to extend our mental simulations.

Nersessian analyzes the role of analogies, images and thought-experimental narratives in our modelling practices (chapter 5). She departs from the standard assumption that these are separate resources, treating them as a continuum of tools for model based reasoning. Instead of direct inferences (mappings from sources to target domains), Nersessian studies how analogies, images and narratives contribute to the creation of intermediary models, with their own sets of constraints, that can be gradually elaborated in a series of representations that finally reaches real world phenomena. The assessment of these intermediary models according to the way they preserve and extend the relevant structural constraints (drawing here on Dedre Gentner’s ideas) provides the epistemic warrant for model based reasoning. Satisfactory models should exemplify features relevant to the epistemic goals of the problem solver (p. 157).

All these claims are illustrated with two case studies of scientific problem solving. In chapter 2 Nersessian revisits her analyses of the historical record of models constructed by James Clerk Maxwell leading to the field equations for electromagnetic phenomena. In chapter 3, she studies an experiment conducted by John Clement in which an expert is asked to solve a physics problem recording every step on his way to the solution. Both cases show at different scales how the cycles of construction, simulation, evaluation and adaptation of models with the aforementioned resources finally yield an original solution. Those interested in bringing together history and philosophy of science will surely appreciate Nersessian’s naturalist approach. Her cognitive appraisal of Maxwell’s models allows her to interpret as positive steps on his way to success what were previously considered misguided attempts at getting there. The analogy between Maxwell’s written records and Clement’s in vivo experiment is equally refreshing. Not every philosopher will enjoy this book though. Nersessian’s approach takes sides and her combination of concept empiricism plus embodied cognition is not precisely mainstream. She is honest enough to admit the limitations of her approach: the view she presents is far from consensual at most points and certainly needs further elaboration. However, the research program she presents is articulated enough to deserve serious consideration. I am not particularly happy though with the purported output of this book: the account Nersessian presents of scientific creativity appears as rather a poor corollary of the previous analysis: if we are able to construct “a potentially infinite number of simulations” many of them will be innovative, but how do we distinguish those of scientific value?

The answer depends of course on the goals and needs of the scientific community. But even if Nersessian wants her account to be social, little is said about how the community chooses the relevant innovations. We are left with the impression that the choice of constraints is always epistemic, but our goals and needs often impose non-epistemic constraints (David Bloor famously argued that scientific concepts are closed according to these latter) and we may legitimately wonder how do they contribute to the success of a model. Part of the plausibility of Nersessian’s analysis depends indeed on her choice of purely epistemic success stories as illustrations. But her argument made me think instead of an already old controversy on failed analogies between physics and economics sparked by Philip Mirowski’s book More Heat than Light in 1989. The equations never proved to be as empirically successful in the target domain (the analysis of the supply and demand) as they had originally been in physics. And yet they were acknowledged as an innovative analogy, perhaps because they observed several constraints considered relevant by some schools of thought in economics. Perhaps for Nersessian the explanation of innovative success and error is symmetrical (in Bloor’s sense), but I would have expected a more explicit discussion of this problem in this book. I can only recommend it nonetheless.



26/7/10

Jan Lauwereyns, The Anatomy of Bias. How Neural Circuits Weigh the Options, MIT Press, 2010.

More and more often practicing scientists from the most diverse fields are writing books for general audiences with a view not only to communicate their own results or the state of the art in their field, but also to draw the more general implications of such findings for, say, our worldview. Whereas the former can be accomplished reasonably well by any competent scientist with a taste for writing, the latter will be more or less engaging depending on what the author has read beyond her discipline. Jan Lauwereyns is a cognitive neuroscientist and a remarkable poet who also enjoys reading across disciplines. And just as in 1621 The Anatomy of Melancholy provided an interdisciplinary survey on its topic, Lauwereyns presents his own anatomy as an "integrative account of the structure and function of bias as a core brain mechanism that attaches different weights to various information sources, prioritizing some cognitive representations at the expense of others" (p. xiv). There is much to praise in Lauwereyns' account, but I wonder to what extent it is really integrative. Let me explain why.

The original core of this book is mostly in the first two chapters, where the author presents the main findings of his own research, which hinge on his version of LATER, the standard model for the analysis of response time distributions generated in visual processing experiments. These experiments measure, on the one hand, the time it takes to a subject (usually monkeys) to convert a sensory stimulus into an eye movement according to the task assigned. On the other hand, they record the level of activity in a neuron (or group of neurons) that code for the relevant stimulus. In LATER the eye movement is understood as a result of a decision, modeled as a function of neural activity: when the function reaches a certain threshold, the eye moves. The function is defined by (1) the starting point of this activity, (2) its slope and (3) the variance of the activity.

In this framework, Lauwereyns claims that bias operates through changes in the first parameter that can be detected in two neural markers. First, the level of activity is significantly stronger in biased neurons than in unbiased neurons at all levels: i.e. for the coded stimulus and for noise alike. But even before the coded stimulus appears, and this is the second marker, the level of activity in biased neurons is also stronger, in a form of anticipatory processing. These superior levels of neural activity make the biased neurons reach the threshold at which decisions are made earlier than biased neurons.

Neural bias should be distinguished from neural sensitivity, the capacity to detect the right signal to act, which is measured by the second and third parameters in the model. The first marker for sensitivity is also a stronger level of activity, but just for the stimulus coded in the neuron, not for noise. The second marker is an enlarged ratio of response to the coded stimulus once it appears, through which the neuron can capture a broader range of signals.

Lauwereyns developed his interpretation of the LATER model namely in order to account for the activity of certain neurons observed in the caudate nucleus of monkeys in an experiment in which they had to make an eye movement to a position where a visual target had been briefly flashed shortly before. Certain caudate neurons, coding for positions where a reward had been obtained in previous blocks of the experiment, showed anticipatory activity before the visual target appeared (what Lauwereyns aptly calls wishful seeing). The monkeys, of course, could not predict where the next flash would come from. Further refinements of this experiment provided evidence that these neurons exhibited a reward-oriented bias of the sort described in the LATER model.

However, in the remaining five chapters, we do not find many more straightforward applications of LATER, but rather an informal examination of other experimental evidence in the light of this model. Hence, in chapter 3 Lauwereyns suggests that there are analogue biases and sensitivity mechanisms for fear (i.e., negative rewards) in the brain. In chapter 4, the author discusses the conceptual compatibility of his conception of bias with two widely studied heuristics in cognitive psychology, whose brain foundations remain as of today unexplored. In the remaining two chapters, Lauwereyns speculates on the more general brain architecture that could support LATER-like information processing. Chapter 5 is perhaps the more daring: Lauwereyns calls for the application of the theory of self-organizing processes to object representations in the brain. In chapter 6, he offers a conjectural model of competition of different neural networks in the brain that could account for evidence gathered in Stroop-like tests for monkeys. Chapter 7 contains the author's musings on the inevitability of bias in our species and how to tame it. The book closes with a Coda on the motivation for the book.

This is, of course, a quite partial summary aimed at capturing what, in my view, constitutes the book's thread or, at least, what I found most original or informative. My major complain about this anatomy is, precisely, how unbalanced it is in every other respect. The author is relatively systematic in the presentation of experimental evidence from his own field, but quite unmethodical in the discussion of everything else. As we have just seen, this is a book in which the author tries to generalize from experiments about certain brain mechanisms of visual processing in monkeys to bias in humans. This generalization requires a number of assumptions that the author clearly acknowledges: to name just one, about our cognitive architecture, how to model it and how to infer it from experimental evidence obtained in different species. Reading the book one gets to know Lauwereyns' views on these particular issues, but there is no introduction to any of them, much less a discussion of the alternatives. The author does not explore in depth positive research on biases in other disciplines, namely cognitive psychology, but rather handpicks examples without presenting their theoretical framework. The final discussion of the social consequences of our biases could have been improved with an examination of, e.g. different policies for fighting conflicts of interests in various domains (do scientific communities really fight biases the way the author thinks, for instance?) I do not think thus that this counts as an integrative anatomy of bias.

The Anatomy of Bias is intended instead as a more personal essay and we get to know more than our usual share about the author's life and tastes. I often got the impression that Lauwereyns digresses just because there is something he aesthetically likes, independently of whether he is right or wrong in the analysis (e.g. his occasional exegeses of Deleuze or Heidegger). Even if I am not particularly happy with this Anatomy, this is certainly a genre worth exploring, and Lauwereyns' attempt deserves all praise for trying to expand the scientific conversation beyond its usual borders.


10/7/10

Alfred Mele, Effective Intentions. The Power of Conscious Will, Oxford, Oxford University Press, 2009

Effective intentions
is a book to be praised by everyone who thinks that philosophers should address issues of public interest, in particular when they arise from the advancement of science. Recent developments in the experimental study of our conscious decisions are challenging some widely spread and deeply rooted intuitions about free will. In Effective intentions Alfred Mele, a prominent philosopher of action, claims that there is no conclusive scientific evidence about the causal efficacy of our decisions and, provided we adopt a proper theoretical framework, there are good grounds to defend the freedom of our will. Mele's book is short and accessible, but it is not popular philosophy: it often engages in scholarly debates and discusses technical points at length. However, those who find the conceptual discussions pervading the popular literature on these topics too rough will probably enjoy reading it. The following summary will provide at least a glimpse of its structure.

In a famous experiment conducted by B. Libet, the participants had to make a movement with their right hand whenever they wished. Libet recorded the electrical activity in the scalp of his experimental subjects together with the activity of the relevant muscles, asking them to signal the time at which they consciously initiated the movement. Libet detected a shift in the activity in the motor cortex that precedes voluntary muscle motion (the "readiness potential") about 550ms before the actual movement took place. The experimental subjects reported the conscious initiation of the movement only 200ms before it started.

Experiments of this sort challenge our common understanding (the folk psychology) of voluntary actions: since we take our will to be free, we would expect the movement to depend somehow on the agent's beliefs or desires. This is why these experiments have captured the popular imagination proving that our will is, in fact, not free. As the title of the book suggests, Mele thinks that intentions are effective and follows a twofold strategy to prove it. On the one hand, Mele reexamines the experimental evidence against free will showing that it is not as conclusive as is often taken to be. On the other hand, the philosophy of action constructed by Mele throughout the past two decades allows him to interpret the experiments in a way that preserve the efficacy of our intentions.

As to the former, Mele puts forward the following interpretation of Libet's experiments: the electrical activity recorded in the scalp 550ms before the action starts would rather be a potential cause of a proximal intention or decision, than any of these two. At least, the electrical patterns associated with the pre-conscious brain activity in Libet's experiments are similar to those recorded in other experiments where there seems to be no apparent unconscious intention or decision. It should be something else, concludes Mele: perhaps some sort of causal input of the intention. In a similar spirit, Mele contests the instances of actions in which intentions apparently have an epiphenomenal role put forward by Daniel Wegner, showing that such actions may not count as intentional. But what sort of intentions could these be?

Mele focuses on ocurrent intentions, defined as executive attitudes towards plans, i.e., being settled on executing them. Such attitude, warns Mele, cannot be reduced to any combination of beliefs and desires. Ocurrent intentions arise from decisions when there is uncertainty about the alternatives; if there is none, ocurrent intentions can be acquired without any explicit decision. Hence, proximal intentions (about immediate actions), at least, need not be conscious: when we act by habit, our acts are no less intentional (we are settled on executing them) even if there is neither a explicit decision nor any awareness of our intentional process. Finally, Mele accepts that intentions may have potential causes and still fully contribute to our actions I hope this very simplified summary will at least suggest why, if we accept Mele's approach, Libet's experiments would not exclude, at least a priori, an intentional interpretation. Our intentions may be considered so despite being causally prompted, unconscious or separated from our beliefs and desires.

However, this does not amount to prove that intentions are causally effective in producing an action. Mele invokes here the evidence on distal implementation intentions: there is evidence showing that people meet non-immediate goals in significantly higher proportion if they are state in advance when, where and how will they achieve them. Prima facie, intentions seem to play a causal role in the explanation of these actions ―and Mele argues at length against alternative accounts in which they do not.

Mele closes the book claiming that science has neither shown that free will is an illusion nor that there are no effective intentions: "this is good news for just about everyone", concludes. But probably "just about everyone" (not this reviewer) will be slightly concerned by the admission that our intentions are causally originated somewhere beyond the realm of consciousness. Mele is quite vague about this point, stating just that our decisions may "more proximally initiate an intentional action that is less proximally initiated" by a potential cause of such decisions (p. 69). I agree with Manuel Vargas (see his piece on Mele's book for the Notre Dame Philosophical Reviews) that this is precisely the point that many would have wanted to see addressed.

It is an indisputable merit of Mele to show that the conceptual framework of Libet's experiments, among others, is often imprecise. Yet, by the same token, it is shown that "the conceptual schemes that we use to interpret and explain our behavior" are equally misleading, which is no less unsettling. Mele's conceptual schemes are certainly more articulate, but this book will not allow the uninitiated reader to grasp them in full. However, unlike many others in philosophy, Mele suggests bits of experimental evidence that could potentially falsify several parts of his theory. We can only hope these tests are actually conducted, making this debate progress in a more empirically oriented fashion, even if the news are not always as good as we once expected.