44th annual lecture series, 2003-04
Was Natural Philosophy in the Late Middle Ages?
Edward Grant, Indiana University
Friday, 10 October 2003, 3:30 p.m.
Frick Fine Arts Auditorium
Abstract: Natural philosophy in the late Middle Ages was the basic curriculum in the arts faculties of all universities, of which there were approximately 75 in Western Europe by 1500. Using Aristotles treatises in natural philosophy, the primary instruments of analysis were reason and reasoned argument and a firm rejection of Biblical and supernatural explanations. The basic format for employing reasoned argumentation took the form of questions. Aristotles natural books were divided into hundreds of questions, such as whether the earth is spherical; whether it is possible for an actual infinite magnitude to exist; whether the existence of a vacuum is possible, and so on. Wherever they were written in Europe, the questions had the same format. Each author posed a question and then defended one of two alternatives. Arguments for the rejected opinion were presented at the outset, and then refuted at the end of the question, after the author had presented conclusions in favor of his own opinion. The questions, and the responses to them, reveal that medieval scholastics departed from Aristotle on many issues, including: the possibility of other worlds, the possibility of motion in a vacuum, and the causes of projectile motion. As the first two questions suggest, medieval natural philosophy was almost as much about counterfactual conditions as it was about the real world that Aristotle described.
Although Aristotelian natural philosophy was abandoned in the seventeenth century, the medieval version of that natural philosophy left a significant, but unacknowledged, legacy to the early modern and modern worlds. The overwhelming emphasis on analytic reasoning to interpret the physical world was the medieval contribution to the scientific advances of the seventeenth century. Nothing as pervasive and extensive had ever previously been attempted in any earlier, or contemporary, civilization. Although much more was neededobservations, experiments, and the application of mathematics wherever required reason is the sine qua non of modern science and the institutionalization of it as embedded in the natural philosophy that was taught and produced in the medieval universities was the first indispensable step in the continuous developments that produced modern science. Without that first step, we might still be waiting for Galileo and Newton.
Not That They Couldn't: Mathematic
Reviel Netz, Stanford University
Friday, 14 November 2003, 3:30 p.m.
2P56 Posvar Halls, Ancient and Modern
Abstract: Why is science not everywhere the same? A popular strategy for answering this question works from what may be called "the argument from conceptual impossibility": that certain authors could not do X because they did not have concept Y (typically, "they" did not have a concept which "we" have). I doubt this strategy. To articulate my doubt, I consider the case of the divide between Greek and later mathematics. In what sense was Greek mathematics non-arithmetical in character, and why?
Infants Little Scientists?
Rethinking Domain-Specificity in Conceptual Development
Jesse Prinz, University of North Carolina
Friday, 5 December 2003, 3:30 p.m.
2P56 Posvar Hall
Abstract: Many psychologists draw a strong analogy between conceptual development and science. They argue that we have innate theoretical principles governing knowledge of9/28/10mals, and other minds). I reject innate ideas and argue instead that infant competence can be explained by appeal to domain-general perceptual biases, memory capacities, and attention systems. I also argue that adult conceptual capacities are perceptually based (not the other way around). In short, I argue that the evidence from development is compatible with an empiricist theory of the mind.
Realism About What?
Laura Ruetsche, University of Pittsburgh
Friday, 16 January 2004, 3:30 p.m.
2P56 Posvar Hall
Abstract: Standard considerations for and against scientific realism can be stated in abstraction from the details of any particular scientific theory. The realist cites a theory T's empirical and explanatory success as grounds for belief in T's truth; her opponent conjures the underdetermination of T by the data T saves to doubt that T gets the unobservable facts right. And so on. In this talk, instead of joining the debate, I will try to figure out what it is a debate about. My strategy will be to see what becomes of standard grounds for and against realism-- what becomes of the realist and anti-realist positions themselves--when a particular empirically successful physical theory is substituted for the placeholder T. The particular theory I will discuss is a theory of ferromagnetism, that is, the phenomenon wherein some substances (for instance, iron) spontaneously magnetize (a symmetry-breaking phase transition) below certain critical temperatures. What becomes of grounds for and against realism, I will try to suggest, is that they start to shake, and start also to seem like the wrong place for a philosopher to be standing, if that philosopher aims to understand the virtues appropriate to physical theories.
Church's Canons: Axioms for Computability
Wilfried Sieg, Carnegie Mellon University
Friday, 6 February 2004, 3:30 p.m.
2P56 Posvar Hall
Abstract: Churchs and Turings theses assert, dogmatically, that an informal notion of computability (or effective calculability) is captured by a particular precise mathematical concept. I will present a conceptual analysis of computability that leads to precise notions, but dispenses with theses. The analysis is embedded in a rich historical and philosophical context.
To investigate computability is to analyze symbolic processes that can, in principle, be carried out by calculators. This is a philosophical lesson we owe to Alan Turing. Drawing on and recasting work of Turing and Robin Gandy, I formulate finiteness and locality conditions for two types of calculators, human computing agents and mechanical computing devices; the distinctive feature of the latter is that they operate in parallel.
The analysis leads to axioms for discrete dynamical systems (representing human and machine computations) and allows the reduction of models of these axioms to Turing machines. Cellular automata and a variety of artificial neural nets can be shown to satisfy the axioms for machine computations.
Populations and Group Selection
Peter Godfrey-Smith, Australian National University and Harvard University
Friday, 19 March 2004, 3:30 p.m.
2P56 Posvar Hall
Warrant and the Value of Truth
Allan Gibbard, University of Michigan
Friday, 16 April 2004, 3:30 p.m.
2P56 Posvar Hall
Abstract: Truth is of value as a guide to choice;
true beliefs guide us in effectively pursuing a wide variety of
aims. Certain true beliefs, it might be thought, are of intrinsic
value too, and the purest of scientific motivations is to learn
the truth for its own sake. I generalize these thoughts to cover
degrees of partial credence. Belief aims at truth, it
is said, and this means perhaps, the following: If a person rationally
values truth and truth alone, the degrees of credence she will most
want to have will degrees of credence that are epistemically rational.
But for many ways one might value truth for its own sake, this fails
to hold. On the other hand, this will hold for anyone who values
truth purely for its value as a guide in the rational pursuit of
other aims. (The theorem that shows this yields, incidentally, a
characterization of payment schemes that would give an expert incentive
to report his degrees of credence truthfully; this problem was posed
by Brier and solved in general by Savage. The payment schemes that
accomplish this are the ones that mimic a possible array of prospective
guidance values.) The upshot raises a puzzle about pragmatic and
non-pragmatic vindications of rationality in belief: Valuing truth
intrinsically would rationally lead to valuing epistemic rationality
only if the way one values truth mimics valuing it instrumentally,
purely for its guidance value.