Dueling Theories: Thought Experiments in Cognitive Science - eScholarship

Dueling Theories: Thought Experiments in Cognitive Science Sam Scott (sscott@ccs.carleton.ca) Department of Cognitive Science Carleton University Ottawa, ON K1S 5B6 Abstract Brook (1999) identified thought experiments as one of the key elements of philosophy's contribution to the cognitive sciences. In this paper, I tackle the question of how and why thought experiments work, and what exactly it is they do for us when they do work. I propose that thought experiments almost always involve two different theories of the world being compared to show that they do, or more often do not, fit together. Sometimes both theories are clearly articulated in the narrative of the thought experiment, but more often one of the two goes unarticulated - the thought experimenter instead relies on our shared folk theories of the world. The strength of some of the more famous and persuasive thought experiments lies in their ability to show that a given theory runs afoul of these deeply held folk intuitions. I will compare the Dueling Theories account of thought experiments to both Brook's empirical account and Brown's (1991) platonic account. Introduction In Hilary Putnam's famous thought experiment, we are asked to imagine a possible world called Twin Earth identical in all respects to our own, but for the chemical composition of what is called water on both worlds. On our world, water is H 2 O, while on twin earth it is something else, say XYZ. Now we are asked to consider what happens when Adam on our world and his counterpart, Twadam on Twin Earth use the word water . Do they mean the same thing? Apparently not. Even though their mental states are the same, the external referent is different. Hence, according to Putnam, meanings just ain't in the head . (Putnam, 1975) In Jackson's (1991) story of Mary the Colorblind Scientist, we are asked to imagine that Mary knows everything there is to know about color. That is to say, she knows everything that can be measured, described, and communicated about what color is and the process by which we perceive it. But she has never actually seen color before. Then one day she sees a color, say red, for the first time and learns something about color that she did not know before, namely what red looks like. The moral we are asked to draw is usually something about the uniqueness and indescribability of phenomenal experience. In introducing these two famous thought experiments, what I am first interested in is how well they work . Reactions tend to vary, but if you're like me, you will be immediately convinced by the story of Mary, but quite skeptical (at least at first) about the Twin Earth story. Perhaps your reactions differ, but the interesting question is why we have the reactions we do. What makes a thought experiment work or not work ? The thesis that I want to defend is that thought experimentation is a meta-activity - a duel between conflicting theories in which one appears to be a clear winner, thus challenging anyone who holds both theories. Thought experiments can never tell us something new about the world, because the world doesn't participate in the experiments. The objects of evaluation are theories of the world. On the other hand, thought experiments are a perfect device not only for revealing problems with various theories of the world, but in some cases, for making clear to us what our theories of the world actually are. It may not be obvious that the two thought experiments described above fit this Dueling Theories story, but if we take the word theory to encompass both scientific and folk theory, then the story not only works but can be quite revealing. Consider Mary first. In one corner, we have some theory in which everything that is part of the physical world can be fully described in scientific terms. In the other corner, we have a folk intuition, based on our own experience of perception, that the phenomenal experience of color is indescribable and would therefore be unknowable to someone not directly acquainted with it. What makes this thought experiment so stunning to so many people is that they may not have realized that they held the folk theory, or that the folk theory was so difficult to give up, until it was put into direct conflict with the other theory. In the Twin Earth experiment, any theory that imparts semantic properties solely to brain states is supposed to lose to a reference theory of semantics. What makes someone embrace or reject Putnam's conclusion will be their pre- existing acceptance or skepticism about whether such reference theories of semantics can be made to work. In what follows, I shall attempt to develop this Dueling Theories view in the context of two alternative views. The first, from Brown's The Laboratory of the Mind (Brown 1991), states that thought experiments can reveal a priori truths about the world through their investigation of Platonic universals. I hope to show that this idea should be rejected, mostly because it actually has very little to offer in aid of our understanding of the nature of thought experiments. The second view, from Andrew Brook's Does philosophy offer cognitive science distinctive methods? published last year in this forum (Brook, 1999), states that thought experiments get part of their value from their empirical content. This empirical content may make thought experiments capable of testing hypotheses against the real world. I believe that with

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.