Chapter 10 Textbook PPT Presentation

Report
Chapter 10 Lecture Notes
Causal Inductive Arguments
Chapter 10
A causal inductive argument is an inductive argument in
which the conclusion is a claim that one thing causes
another. (286)
For example:
(i) Clogged arteries cause heart attacks
(ii) A rough surface produces friction
(iii) Exercise during heat causes sweating
Causal inferences are often an attempt to explain or predict
an outcome.
Chapter 10
Cogent arguments in support of causal claims are difficult
to come by, and we should take great care when dealing
with causal premises. There are lots of ways to express
a cause in English without using the word ‘cause’ and
here are just a few ways from pages 286-7.
C produced E
C brought about E
C created E
C influenced E
E was determined by C
E was induced by C
C was responsible for E
C led to E
C affected E
As a result of A, E occurred
E was the result of C
E was an effect of C
Chapter 10
The meaning of cause can vary as well because of the
different nature of causal relationships. Here are four
possible meanings for cause:
(i)
(ii)
(iii)
(iv)
As a necessary condition
As a sufficient condition
As both a necessary and sufficient condition
As a contributory factor (neither necessary or sufficient)
Each of these meanings of ‘cause’ are useful in analyzing
arguments and premises in argument. We will see
examples of each.
Chapter 10
Correlation vs. Causation
Correlation is a symmetrical relationship, which causation
is asymmetrical. Here are the three ways we might
classify correlation relationships.
(i) Positive correlation: if a higher proportion of Qs than non-Qs are H, then
there is a positive correlation between being Q and being H.
(ii) Negative correlation: if a smaller proportion of Qs than non-Qs are H, then
there is a negative correlation between being Q and being H.
(iii) No correlation: if about the same proportion of Qs as non-Qs are H, then
there is no correlation between being Q and being H.
A significant correlation is one that is reliable. (288)
Chapter 10
When Q is positively correlated with H, then one of the
following has to be true, but it doesn’t have to be a
causal relationship.
1.
2.
3.
4.
Q is a cause of H.
H is a cause of Q.
The positive correlation of Q and H is a coincidence.
Some other factor, X, is a cause of both Q and H.
It should be clear now, why we cannot conclude that
correlation implies causation given these four
possibilities.
Chapter 10
Associations and Links
To say that two things: A and B are linked is to claim that
more than just a correlation between A and B exists.
Linking suggests that there is a causal relationship
between A and B.
Linked seems to have become a code word for unknown
causal relationship. Be skeptical and investigate the
data and reasoning involved with such claims.
Chapter 10
Mill’s Methods for Causal Reasoning
John Stuart Mill created two methods for finding causal
relationships and one joint method.
The Method of Agreement is a test for a necessary condition.
Any factory that is absent when G is present is eliminated as a possible necessary
condition of G.
The Method of Difference is a test for a sufficient condition.
Any factor that is present when G is absent is eliminated as a possible sufficient
condition of G.
The Method of Agreement and Difference is a joint test for necessary and sufficient
conditions. Review pages 296-298 for detailed explanations.
Chapter 10
Inference to the Best Explanation (IBE)
The general form of an IBE is the following:
1. D exists.
2. H 1 would explain D.
3. H 1 would offer the best explanation of D.
Therefore, probably,
4. H 1.
Where D stands for data and H1 stands for a hypothesis.
We call IBEs abductive arguments following the
language of C.S. Peirce. Abductive arguments lead to
an explanatory hypothesis according to Pierce.
Chapter 10
When evaluating IBE arguments, the third premise needs
support via a subargument and we can detail it this way.
(a) Plausibility
1. Empirically adequate
2. Probable
3. Not ‘ad hoc’ (for this purpose hypothesis”
(b) Falsifiability – means makes a genuine assertion that is
compatible with some data an incompatible with other
data.
Now what makes one explanation better than another.
Chapter 10
Here are some considerations for comparing explanations:
(1) More plausible than the alternatives
(2) Explains more/has more explanatory power
(3) Simplicity – the explanation is more simple
These condition in conjunction with plausibility and
falsifiability are what give rise to cogent, inductive IBE
arguments. But we should be careful with IBE
arguments because of how we support the third
premise.
Chapter 10
Errors and Fallacies in Causal Reasoning
The Post Hoc Fallacy – this is sometimes called the post
hoc ergo propter hoc fallacy. The full phrase means:
“after this, therefore because of this” And it is a causal
inference fallacy. (304)
Just because one thing comes before another does not
mean that the first thing was causally relevant to the
second, and this is just what the post hoc fallacy claims.
It is classic bad reasoning.
Chapter 10
The fallacy of objectionable cause: occurs when a person
argues for a causal interpretation on the basis of limited
evidence and makes not attempt to rule out alternative
explanations of the event. (306)
Sometimes this fallacy is called the false cause fallacy.
This kind of fallacy happens a lot in election debates and it
is similar to a confusion of correlation with causation.
Chapter 10
Begging the Question in a Causal Account:
This is just a case of begging the question in a situation
where there is some causal claim our account in
question. So, a person assumes the conclusion or
something logical equivalent to the causal conclusion as
support for it.
See page 308 for a detailed example.
Chapter 10
Causal Slippery Slope Arguments:
Causal Slippery Slope fallacy claims in the premises that
some action would be wrong because it would let off a
series of side effects ending ultimately in general
calamity.
Causal slippery slope fallacies can also go in the other
direction claiming that something would be good
because it would give rise to certain good effects.
Fixing these fallacious arguments generally requires
providing cogent subarguments for the premises.
Chapter 10
Terms for review:
Abductive argument
ad hoc hypothesis
Causal inductive argument
Causal slippery slope
Correlation
Explanatory Power
Fallacy of objectionable cause
Falsifiability
Inference to the Best Explanation (IBE)
Plausibility
Post hoc fallacy
Simplicity

similar documents