Unit 3: Goodman, Zero/Sainsbury, Kant Flashcards
1
Q
a formal theory of confirmation
A
- a formal theory of good inductive reasoning
- how do particular repeated observations confirm a generalization?
- the idea that relations of confirmation might be modified in style of deductive logic
2
Q
Formal theory of confirmation…deductive to inductive
A
- example of deductive logic
- all humans are mortal
- socrates is human
- C: socrates is mortal
- notice, we only had to refer to the form and not the actual content
- attempt to argue that inductive relations and reasoning or arguments might be given as a formal theory as well
- Goal: to formulate inductive logic like deductive logic, borrowing deductive logic whenever possible
3
Q
deductive
A
logical reasoning
4
Q
inductive
A
evidential reasoning
5
Q
the ravens problem
A
- observation: repeatedly observe black ravens
- hypothesis: all ravens are black
- logical equivalence/contrapositive:
- if its a raven, then its black if and only if for all x if not black it is not a raven
- Problem: all ravens are black, if and only if all non black things are not ravens
- evidence for hypothesis that all nonblack things are not ravens
- swan f is always white, something not black, not raven evidence must also support its logial equivalent
- so, swan f that is white, confirms all ravens are black
6
Q
goodman problem
A
- “indoor orinthology”
- instances of generalization confirm the generalization of its equivalent
- ridiculous, because we can’t investigate the color of ravens without going outside
- ie brown shoe example
- induction should require some observation
7
Q
what the raven’s problem shows
A
- PGS reports a point made by I.J. Good: trouble because comparing one proposition (oberservation) with another (hypothesis)
- a binary analysis of support relations
- a 3 place relation can exhibit disconfirmation
- “confirmation depends on other factors”
- PGS reports another point made by Good
- ordinality and the possibility to confirm or disconfirm
- “depends on the order in which you learn of the properties of an object”
- ordinality and the possibility to confirm or disconfirm
- *Overall the problem shows
- ordinality and the possibility to confirm or disconfirm
8
Q
the new riddle of induction intro
A
- formal theories of confirmation and instance confirmation
- deductive: universal instantiation
- all Fs are Gs
- A is an F
- C: A is a G
- inductive: instance, enumerative
- emerald a is green
- emerald b is green
- C: all emeralds are green
- model of FTC of good inductive reasoning after deductive reasoning
- we learn something about the nature of support
- goodman undermines the prospects for not developing this idea
- goodman is NOT arguing that confirmation is impossible or that relations of confirmation dont exist
- hes attacking the idea that a purely formal theory can be successful, there can never be a formal theory of induction and confirmation
- characterized with variables, not with content
- goodman is NOT arguing that confirmation is impossible or that relations of confirmation dont exist
9
Q
“grue”
A
- an object is grue IFF it was first oberserved before 2015 and was green, or 2 it was not first observed before 2015 and is blue
- argument is not good because it says emeralds we observe from now on will be blue
- but this argument has exactly the same form as the good inductive argument we saw before
- All observations of emeralds before 2015 confirm both hypotheses
- 1: any yet to be seen emerald is green
- 2: any yet to be seen emerald is blue
- Therfore, FTC does not seem to be possible
10
Q
the new riddle of induction & suggestions
A
- q: what exactly is wrong with bad inductive arguments? what makes bad induction bad?
- Suggestions to fix
- restrict such words as grue because they refer to time
- properties words should refer to should be natural kinds (but what is outside of chemical elements?)
- dont employ complex terms, replace with simpler ones
- Brings us to: the curve fitting problem
- do we choose a straight line to fix the curve because its simpler? then it would be wrong!
- Brings us to: the curve fitting problem
11
Q
Zeno’s background and 2 radical conclusions
A
- Background
- use of conceptual clarification of calculus and the development of rigorous methods for invetigating the infinite
- challenging fundamental concepts in metaphysics
- the relation of the parts of an object to the whole object, mereology
- problems about the way objects persist through time, persistence
- the nature of space, metaphysics
- abstract qs about exactly how math or geometric concepts can be applied to material objects, from metaphysics to physics
- Claim 1
- there is not more than one thing
- claim 2
- change or motion is not possible
12
Q
spatial relations and mereology
A
- spatial relation and parts whole relations
- example: how can 1 thing, a book, be many things: pages, molecues
- answer: its not, the book is not identical to many things, the book just has many parts
- zero attempts to show that the idea that material objects have parts is absurd
- so, we must reject the claim that anything has parts–it is all one thing
- example
- 1: if a book has parts, infinitley many
- 2: if a book has infinite parts, it is infinitley large
- 3: so, if a book has parts it is infinitlyey large
- 4: no book is inifinitley large
- C: no book has parts
13
Q
Re premis 1: if a book has parts, infinitley many
A
- if a book has parts, then the region of space it contains has parts, and the book has a part corresponding to every part of space it fills
- so, if a region of space has parts, it has infinitley many parts
14
Q
Re premis 2: if a book has infinite parts, it is infinitley large
A
- if a region of space has infinitley many parts, it is infinitley large
- this claim does a lot of work for the argument
- parthood for space requires infinite divisibility
- reasonable, because finite amount of space parts is finitley large
- there is something wrong with this
- a finite region of space contains infinitley many parts
- Q: how do they all fit?, thet are continually getting smaller
- Distinction
- **if, for some finite size, a whole contains infinite parts, none smaller than this size, then the whole is infinitley large
- true: saying 0.5 + 0.5…=infinity (not going to be a finite size)
- if a whole contains infinitley many parts, each of some finite size, then the whole is inifinitley large
- False: they can be smaller than the other ie 0.4 + 0.3 + 0.2…does not equal infinity
- **if, for some finite size, a whole contains infinite parts, none smaller than this size, then the whole is infinitley large
- so, there is no contradiction in the idea that space is divisible. but this does not entail that space, the actual space, actually is infinitley divisible
- reasonable, because finite amount of space parts is finitley large
- premis 2 is false: zenos argument against plurality fails, fails because of of this statement **
- for all we know, space might be granular
- zeno has not given us any reason to think that space is granular
15
Q
The Racetrack, the dichotomy, sainsbury
A
- paradox: motion is impossible
- traveling from a to b requires infinitley many journeys
- it is impossible for anything to complete infinite number of journeys
- C: impossible to get from A to B
- C2: points are arbitrary, so all motion is impossible (False)