Friday, April 29, 2011

Test for the “Human”

Christine Rosakranse
Response for 369 - Nass
Test for the “Human”

    Any test for humanness is a difficult one because the definition of human is different depending on which aspect of the individual you are studying.  Of course, there is the biological component of a human being, and that is well-defined.  We have certain DNA and the expression of those genes leads to common physical traits.  Mutations do occur and there are outliers of the system, but they lie well within a range of possibilities.
    However, the ontological essence of humanity remains out of the scope of the Turing test.  He relies on transmitted communications in order to remove the face-to-face interaction, but that would be exactly where humanness lies.  To be honest, I would argue that humanness is a spectrum.  On the one end, you have a rock, inanimate and unfeeling.  At the fully human end, you have an entity that hopes, worries, loves, and that is capable of engaging with other members of humanity. 
    I always think of Data from Star Trek: The Next Generation when this question arises.  He progresses through the seasons trying to become more “human”, playing chess, painting, and even have conjugal relations with another crew member.  But throughout all of that practice, he submits that he is not human.  In the movie, he gets the emotion chip, one of my favorite deus ex machina devices ever.  They do not explain how it works.  They just pop it in and boom - manic, all-feeling Data is born.  I would say that at this point he became human, and would pass any Turing or Nass test.
    This is not to say that anything with emotions is human.  I know my dog worries a lot about treats, and the scarcity of treats, and all beef related items in the world.  But it’s how one worries that makes a thing human or not.  On my spectrum, though, a dog would be more human than a rock, less human than me, but not because she can’t use grammar. 
    Perhaps my test would be: To what extent does the entity have the ability to create an explanatory narrative?  We constantly distinguish “what matters”, using a combination of logic and very nearly arbitrary rules.  I think a computer wouldn’t be able to say which is more important, a rainbow or a cricket.  And a human would, coming up with some narrative, that may or not make sense to a computer, but that would make sense to another human being.  

Thursday, April 28, 2011

What is science?

Kuhn and Lakatos

    Kuhn’s account of paradigm shifts describes anomalies and inconsistencies as their root.  This argument suggests a single model for change.  Whether or not Kuhn believes this to truly be the only model would definitely color my ability to believe in his logic.  His definitions, I find also, are too restrictive. 
    He further explains that, as demonstrated by astronomy, the holes in a theory serve as a latching point for a new theory, without which it is unlikely for a new theory to hold.  Retooling is an expensive process (intellectually and otherwise).  Not just any theory would be considered for study.  Something must be the catalyst for change.
Another way to state this is that “crises are a necessary precondition for the emergence of novel theories” (pg. 77).    This is a very logical way to think about these changes.      However, when he writes “To reject one paradigm without simultaneously substituting another is to reject science itself” (pg. 79), I cannot agree with him and neither can Lakatos.  Suggesting that you work within one paradigm until you create a new one once again calls to the no-interregnum argument against Kuhn’s formation.  And while he does admit to a possible period of confusion, he does not call this science. 
    Lakatos, on the other hand, counters this by first introducing a concept from Popper that “all theories are not only equally unprovable but also equally improbable” (pg. 95).  Dogmatic falsificationism assumes a “psychological borderline” between theory and factual prepositions.  Once again, psychology comes into the realm of discourse because science can never be a “pure” exercise free from human foibles and intentions.   “There is no natural (i.e. psychological) demarcation between observational  and theoretical prepositions” (pg. 113).  He instead changes a paradigm from solid to “background knowledge.”  
    One interesting side note comes from a footnote which states that some might see “methodological falsificationism as an existentialist philosophy of science” (pg, 113, footnote 2).  Adding sophistication to methodological falsificationism involves distinguishing between interpretive theory and explanatory theory (pg. 129).  Lakatos then replaces the notion of a single theory as being scientific with the necessity for a succession of theories to be scientific.  I think that this is a better representation.  Taking Kuhn’s view of science-no science-science to explain periods of confusion does not account for the intricacies of science and for the entire process. 
    Lakatos uses the word “metaphysics” which represents an interesting focus for discussion, which in this case describes methodological rules as being “....motivated by the metaphysical principle that any explanation is only approximate because of the infinite complexity of the factors involved in determining any single event.”

So two questions we could ask are:  To what extent does seeing methodological falsificationism as an existentialist philosophy of science affect the researcher?  Does seeing science in a succession of theories make more sense than claiming individual theories as scientific?