Laudan on Underdetermination

Laudan's classification of kinds of underdetermination is worthwhile, and provides a useful framework in which to think about such claims for the rest of the semester.

On the other hand, his characterizations of the views of his opponents, especially, are wrong to the point of being embarrassing. I'll just point out the most egregious example: Laudan says (84):

Quine has claimed that theories are so radically underdetermined by the data that a scientist can, if he wishes, hold on to any theory he likes, "come what may."

His evidence is given in footnote 12, where he says,

Quine specifically put it this way: "Any statement can be held true come what may, if we make drastic enough adjustments elsewhere in the system of belief."

---

Laudan's classifications: An underdetermination thesis can be classified along the following dimensions:

• Descriptive–normative. A norm is a standard or rule. "Norm" has an almost moral connotation. A descriptive thesis is one about what people actually do. A normative thesis is one about what people should appropriately do. (In the normative case, we should ask, what is the operative norm. Laudan always considers something like Scientific Rationality.)

* Deductive–ampliative. A deductive argument is one in which if the premises are true, then the conclusion must also be true. An ampliative argument "amplifies" the premises: the premises give good reason to believe that the conclusion is true. A set of evidence deductively underdetermines a conclusion if it is consistent with the evidence that the conclusion could be false—there is no contradiction involved. A set of evidence ampliatively underdetermines a conclusion if it does not give good reason to believe the conclusion. There can be many kinds and degrees of ampliative confirmation. Laudan lists (91)

• • be logically compatible with the evidence (deductive)
  • logically entail the evidence
  • explain the evidence
  • be empirically supported by the evidence

• Nonuniqueness$ndash;egalitarianism. A "nonuniqueness" form of underdetermination is one that claims that a body of evidence does not uniquely determine a theory (there are at least two theories left on the table by the evidence). An "egalitarian" form of underdetermination claims that the body of evidence leaves all theories in an equally good state.

A claim of underdetermination could be made, I suppose, for a single body of evidence, but what Laudan has in mind are claims for bodies of evidence of two different types:

• The finite body of evidence we already have–a total body of evidence.

That distinction is made by Laudan, but not named.

There is another distinction I want to add:

• Universal—possible underdetermination. A universal underdetermination claim is one that claims that the evidence always underdetermines the theory. A possible underdetermination claim is one that claims that the evidence may sometimes underdetermine the theory.

---

What kinds of underdetermination claims are correct?

Descriptive, strong (empirically supported), nonunique, total (?) evidence, possible: The flat earthers show that this is possible.

Normative, ampliative, nonunique, total evidence, universal: Mereology shows this.

Normative, ampliative, nonunique, finite evidence, possible: Atomism vs continuous matter.

Normative, deductive, "egalitarian," finite evidence, universal: Any amount of finite evidence is deductively compatible with many universal claims of the sort made in theories

If a conclusion is ampliatively underdetermined by evidence, it is certainly deductively underdetermined.

-- ShaughanLavine - 17 Feb 2009