|
|
|
|
| 21
February
(Wednesday) 5:00
PM Room 226 |
| Gábor
Hofer-Szabó |
Institute of
Philosophy, Research Centre
for the Humanities, Budapest
|
| |
|
Commutativity,
commeasurabilty, and
contextuality
|
Kochen-Specker theorems,
allegedly proving quantum
contextuality, are based on the
assumption, standard in textbook
physics, that commuting operators
represent commeasurable
(simultaneously measurable)
observables. However, in all
versions of the Kochen-Specker
theorem there are commuting
operators which can hardly be
interpreted as representing
simultaneously performable
measurements. Lacking
commeasurability, two extra
assumptions are needed to get the
contradiction, the violation of
neither of which, however, implies
contextuality of any sort.
|
| 28
February
(Wednesday) 5:00
PM Room 226 |
| Maria
Kronfeldner |
Department of Philosophy, CEU,
Budapest
|
| |
|
Two
reasons why the
purist approach to
science fails with
respect to thick
concepts
|
‘Thick concepts’ are
value-laden, i.e., they have an
evaluative and a descriptive
aspect. The concept of aggression
is a standard example. When a
psychologist writes that “boys are
more aggressive than girls” then
this is not a purely descriptive
claim. Purists claim that
scientists can and should purify
thick concepts, by establishing
(what I call) ‘naked numbers,’
i.e., an operationalization, e.g.
an A-index that objectively
measures the respective behavioral
trait meant with the term
‘aggression’. This talk will
present two problems and thus two
reasons why the purist approach
must fail with respect to thick
concepts. (1) The outsourcing
problem: Reducing thick concepts
to naked numbers can be rejected
as a solution for making science
value-free since the connection to
the original conception and thus
the values involved (e.g. what lay
people call aggression) needs to
be made somewhere in the
process of knowledge production,
so that the knowledge produced
stays socially significant, i.e.,
provides reasons for action.
Purification turns out to be inefficient.
(2) The under-determination
problem: The conceptual decisions
in reducing thick concepts to
naked numbers cannot be made
objectively (i.e., completely free
of bias) because an
operationalization is always
underdetermined by data. In
effect, different
operationalizations let one see
different evidences: whenever a
phenomenon is reconstituted and
its definition re-operationalized,
some evidence can be made to
disappear. Such
disappearance of evidence will be
shown to be necessarily biased and
often value-laden, by analyzing
the history of research on
aggression. Purification turns out
to be impossible.
|
|
|
|
