HPS 2501/Phil 2600 Philosophy of Science Fall 2011
Back to course documents.
(a) You are required to submit the small warm up exercise and then your choice of THREE only of the short papers described below.
(b) Optional additional assigment: you can, if you wish, submit one additional short paper. The paper grade will be determined from the four best papers submitted, including the warm up exercise.
(c) The short papers are intended to be short--500 to 1000 words--but rich in content. In counting the words, footnotes are included, but not the list of references.
(d) The papers are due one week after the last paper of the relevant section has been read in class.
(e) Papers will be graded according to how well they clearly state a thesis in philosophy of science and argue cogently for it. See policies on papers.
(f) Remember also my policies late submission. (Don't.)
(g) Send the paper to me in email (firstname.lastname@example.org) in an editable file, that is, not a pdf. I find rtf = "rich text format" the most flexible and generated by most word processors. Microsoft "doc" files are OK too.
Leonard Leibovici's "Effects Of Remote, Retroactive Intercessory Prayer On Outcomes In Patients With Bloodstream Infection: Randomised Controlled Trial" (British Medical Journal, Vol. 323, No. 7327 (Dec. 22 - 29, 2001), pp. 1450-1451) and the subsequent exchanges of letters (British Medical Journal, Vol. 324, No. 7344 (Apr. 27, 2002), pp. 1037-103)9 represents an extraordinary episode in medical research. The double blind study found that prayer for recovery could act retroactively, that is, into the past. The study found that prayer for patients in the year 2000 was associated with more rapid recovery of patients that had already happened in the preceding decade. The paper concludes that "retroactive, intercessionary prayer...should be considered for use in clinical practice."
The study raises many foundational issues in philosophy of science. Are there some results that are somehow beyond the realm of proper scientific study? If a result is demonstrated by a properly constructed controlled trial, can its results be overridden?
The warm up exercise is to assert a non-trivial claim in philosophy of science concerning the trial and argue for it.
This is an exercise in the stating of a clear thesis and arguing cogently for it. This is not the place for any rambling discussion. Get straight to business. State the thesis clearly and lay out the argument. Don't bury it in chatter. You have 500 words maximum. Use them well.
CHOOSE THREE OF:
Proponents of the older "syntactic" and the newer "semantic" view of theories battled fiercely. Which, if either, has the better characterization?
Analogical models and idealizations purposefully introduce falsehoods as, supposedly, a necessity of tractable theorizing. If they are eliminable, why not eliminate them? If they are ineliminable, how can we ever know which results of our theorizing are merely artifacts of these falsehoods?
According to inductive generalization, things that obtain some of the time (this A is B) support the hypothesis that they will obtain all of the time (all As are B). This scheme supposes a uniformity of nature. However nature is uniform in virtually none of its aspects. Why has this venerable form of inductive inference been so successful?
According to the scheme inference to the best explanation, a theory that explains the evidence is more likely to be true. What has successful explanation to do with truth?
Is there a singe correct theory of induction? If so, which is it? If not, does the multiplicity of theories suggest the basic idea is confused?
Bayesian confirmation theory assigns a probabilty, a definite number, to the strength of support, even when intuitively the support is not so precisely expressible. Should we take this as a precise correction to our imprecise intutions? Or does it show that the Bayesian theory needs to be diluted?
Is the interpretation we give to probability important for Bayeisanism?
We do not teach creationism in schools because it is not science. We do not treat patients with homeopathic medicine because homeopathy is not scientific. We seem to know when something is science or non-science. Yet our efforts to provide a demarcation criterion have failed.
How can we advise the larger public grappling with practical issues of creationism, homeopathy and more?
We have seen various skeptical attacks on induction ("The" problem of induction; "grue"; "underdetermination.) These skeptics generally agree that our belief in science is due to some arational process, such as habit or yielding to social pressure. If we accept this verdict, how can we explain the continuing success of science? If we do not accept it, how can we answer the skeptics' arguments?
Answer in the context or one or more of the problems of induction listed.
Is there a singe correct theory of explanation? If so, which is it? If not, does the multiplicity of theories suggest the basic idea is confused?
It is generally accepted that a successful science must explain the relevant phenomena. Conforming to them and even predicting them in minute detail, is not enough. Why is it not enough?
Is explanation an essentially subjective process through which we achieve a psychological sense of understanding? Or is it an objective relation within science?
Antirealist doubt that science has the depth of knowledge of nature naively attributed to it; but they must explain the success of science. Realists revel in the extraordinary successes of science; yet they must explain our now standard expectation that the latest success will be overturned by the next revolution.
What is your view of this tension?
Interventionist accounts of causation require a prior causal notion of intervention or manipulation, in order to discern whether something really is a cause of something else. Does this make the account circular?
While causation seems to figure centrally in every science, we seem unable to generate a universally admissible account of its nature. Why is this?
A provocation: Chemical systems are just very complicated phycisal systems. Biological systems are just very complicated chemical systems. So should we not expect that in principle the lower level theory (physics, chemistry) can explain the higher level theory (chemistry, biology)?
Experiments, in which one intervenes and sets up artificial situations, are widely accepted as the best grounding for a science. There are sciences, such as astronomy or archaeology, in hich interventions are largely impossible. As a result, are they are lesser sciences or less well grounded? If not, is experiment overrated?
Hacking's much quoted remark is that "Experiment has a life of its own..." (Representing and Intervening, p.xiii) It is a metaphor. Experiments are not alive. Express the claim made directly without the metaphor. Is it correct?
The era of writing the grand account of scientific change--paradigms and revolution, research programmes, progress and problems--has passed. There is now a growing sentiment that it was far too ambitious in seeking sweeping statements about the essential nature of all science. Was it too ambitious?
Kuhn listed values that play a role in theory choice (accuracy, simplicity, consistency, scope, fertility). Values, as opposed to facts, are freely chosen. Does this mean that our theory choices are arbitrary?
Note added at the end of term:
Oops! This last question does not capture what I intended to ask. Here is what I should have asked and also for a speech on why I take a dim view of Kuhn's "values" paper.