Philosophy of Science

Questions for the Mid-Term Examination

Fall, 2006

Curtis Brown

The exam will consist of two parts. The first part will contain several (perhaps 6-8) questions asking for brief descriptions of some of the key ideas we have discussed. The second part of the examination will ask you to write more substantial essays on one or two broader questions. Answers to the essay questions will be graded on (1) the degree of familiarity with, and understanding of, the details of the readings exhibited by your answers; (2) quality of philosophical argument, and degree to which you consider and respond to views opposed to your own; (3) clarity of organization and expression. The outline of the material we have covered this semester on the web site may help you to organize your thought for the examination. Of course, you should also review the readings! The overview does not contain enough material to enable you to do well on the exam without also consulting the texts!

Possible short-answer questions:

Possible Essay Questions:

1. A November 5, 1998 article in Nature, titled "Jefferson fathered slave's last child," argued that genetic testing showed that Thomas Jefferson fathered the last child of his slave Sally Hemings.  A recent article in the online journal naturalScience includes the following paragraphs:

In truth, however, the study proves very little. As Gary Davis of
the Evanston Hospital commented in a letter to Nature (6), “any
male ancestor in Thomas Jefferson’s line, white or black, could
have fathered Eston Hemings. Plantations were inbred
communities, and the mixing of racial types was probably common.
As slave families were passed as property to the owner’s offspring
along with land and other property, it is possible that Thomas
Jefferson’s father, grandfather or paternal uncles fathered a male
slave whose line later impregnated another slave, in this case Sally
Hemings.”

And that does not exhaust the possibilities. As Samuel Francis
wrote in the New American “There seems to be little doubt that
Thomas did share the distinctive Y chromosome found in the
present-day descendants of his uncle, but so did the uncle and all
his descendants down to the present day. So did Thomas
Jefferson’s brother Randolph, as well Randolph’s six known sons.
How are we to know that Field Jefferson himself, or one of his
sons (or, for all we know, illegitimate sons as well), or one of their
descendants, or Randolph, or one of his six legitimate sons (or
illegitimate sons perhaps), or one of their descendants was not the
father of Eston Hemings or one of Eston’s male descendants at
some time during the last 200 years?” (6).

Discuss this example in light of the H-D and Bayesian approaches to confirmation. How would the H-D approach analyze this example? Are there limitations or drawbacks to the H-D account of the example? How would a Bayesian analyze the example? (Set the example up in such a way that Bayes's rule applies to it. What is the theory being tested? What is the new evidence? On what basis does the article argue that the new evidence does very little to increase the posterior probability of the theory? You don't need to assign specific probabilities, but explain in general terms what Bayes's rule suggests about this example, thinking about the example simply in terms of which probabilities are high and which are low rather than assigning precise numerical figures.) In what respects does the Bayesian approach seem illuminating? Are there drawbacks or disadvantages to this approach?

(By the way, this isn't germane to the exam, but in case you're interested, the same journal also published a follow-up letter with some additional historical information.)

2. Give a careful explanation of the covering-law model of explanation. (This is a general term that includes both Hempel's Deductive-Nomological and his Inductive-Statistical models.) What conditions must be met for an explanation to be satisfactory, on this view? Explain the difference between objections that these conditions are not necessary and objections that they are not sufficient. Explain and evaluate one objection of each sort. Conclude with an overall assessment of the covering law theory.

3. Explain Bayes's Theorem. Discuss the relation between Bayes's Theorem, the H-D model of confirmation, and Popper's falsificationism.  Conclude with an overall assessment of the usefulness of Bayes's Theorem in thinking about issues of confirmation. Your assessment should include your evaluation of criticisms of "Bayesianism" (as discussed in Curd & Cover's "Commentary" and briefly summarized at the end of my handout on Bayes's Theorem) as well as supporting arguments.

4. Explain the Paradox of the Ravens. Show why it is a puzzle for the model of confirmation by positive instances, and also for the H-D model. Then explain and evaluate a Bayesian response to the paradox.

5. Consider two examples of purported pseudosciences (any two you like; possibilities include astrology, alchemy, parapsychology, sociobiology, psychoanalytic theory, Pyramidology, etc. Of course, some of these have better claims than others to scientific respectability!). Explain Popper's criterion of demarcation, and apply it to your examples. Explain how Kuhn's criterion differs from Popper's, and apply it to your examples as well. Finally, consider Ruse's suggestion about how to distinguish science from non-science and apply it.  Conclude with a discussion of the relative merits of the three strategies for distinguishing science from pseudoscience.

6. Explain the "flagpole and shadow" objection to the covering-law model of explanation. Then consider van Fraassen's story "The Tower and the Shadow."  Explain the relation between this story and the objection. How might van Fraassen use the story to argue that his own pragmatic account of explanation is superior to the covering-law model? 

7. Consider the following interesting fact: when performance on graduate admissions examinations is compared by academic major, philosophy majors are the only group to score significantly better than average on all four of the following: the GMAT, the LSAT, the verbal portion of the GRE, and the quantitative portion of the GRE. Philosophy majors had the highest average on the verbal GRE, second highest (after mathematics) on the GMAT, and third highest (after mathematics and economics) on the LSAT.

Does the fact that these students are philosophy majors explain their excellent performance on these examinations? Consider the following I-S model. General law: Philosophy majors are more likely than other majors to do well on the GRE. Initial condition: Alfred is a philosophy major. Explanandum: Alfred did well on the GRE. Is this a good example of an I-S explanation? (Does it satisfy Hempel and Oppenheim's criteria?) What additional constraints are imposed by Ruben's causal model of explanation, and how do they apply to this case? Compare the virtues and vices of these models of explanation with particular reference to this example.

8. Discuss the metaphysical issue of scientific realism. Compare the views of Maxwell, van Fraassen, and Arthur Fine. Explain which of these views you find most adequate, and why.



Last update: October 16, 2006
Curtis Brown | Philosophy of Science | Philosophy Department | Trinity University
cbrown@trinity.edu