What do you think?
Rate this book
Causation: A User's Guide
Causation is at once familiar and mysterious. Many believe that the causal relation is not directly observable, but that we nevertheless can somehow detect its presence in the world. Common sense seems to have a firm grip on causation, and much work in the natural and social sciences relies on the idea. Yet neither common sense nor extensive philosophical debate has led us to anything like agreement on the correct analysis of the concept of causation, or an account of the metaphysical nature of the causal relation. Contemporary debates are driven by opposing motivations, conflicting intuitions, and unarticulated methodological assumptions.
A User's Guide cuts a clear path through this confusing but vital landscape. L. A. Paul and Ned Hall guide the reader through the most important philosophical treatments of causation, negotiating the terrain by taking a set of examples as landmarks. Special attention is given to counterfactual and related analyses of causation. Using a methodological principle based on the close examination of potential counterexamples, they clarify the central themes of the debate about causation, and cover questions about causation involving omissions or absences, preemption and other species of redundant causation, and the possibility that causation is not transitive. Along the way, Paul and Hall examine several contemporary proposals for analyzing the nature of causation and assess their merits and overall methodological cogency.
The book is designed to be of value both to trained specialists and those coming to the problem of causation for the first time. It provides the reader with a broad and sophisticated view of the metaphysics of the causal relation.
A User's Guide cuts a clear path through this confusing but vital landscape. L. A. Paul and Ned Hall guide the reader through the most important philosophical treatments of causation, negotiating the terrain by taking a set of examples as landmarks. Special attention is given to counterfactual and related analyses of causation. Using a methodological principle based on the close examination of potential counterexamples, they clarify the central themes of the debate about causation, and cover questions about causation involving omissions or absences, preemption and other species of redundant causation, and the possibility that causation is not transitive. Along the way, Paul and Hall examine several contemporary proposals for analyzing the nature of causation and assess their merits and overall methodological cogency.
The book is designed to be of value both to trained specialists and those coming to the problem of causation for the first time. It provides the reader with a broad and sophisticated view of the metaphysics of the causal relation.
- GenresPhilosophyMetaphysics
300 pages, Paperback
First published January 1, 2013
4 people are currently reading
191 people want to read
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 3 of 3 reviews
December 5, 2018
I'm very conflicted about this book. On the one hand, it does an excellent job of providing an overview of several different contemporary theories of causation and demonstrating whether and how they can overcome a wide variety of problems. The authors, Paul and Hall, also spend a significant amount of time discussing the methodology they are using (e.g. counterexamples by cases), guiding ideas in the metaphysics of causation (e.g. reductionist versus nonreductionist accounts), and what they think are particularly fundamental ideas that we have about causation (e.g. that causation is, in some important sense, intrinsic). This is all good.
However, there are several problems with the book: an occasional lack of clarity and occasional mistakes. Regarding the first problem, sometimes Paul and Hall just did not explain themselves clearly. For instance, on pages 81-82 they state what they claim is Hitchcock's proposal of de facto causation: where C consists in some variable taking on a certain value, and E likewise consists of another variable taking on a certain value, C is a cause of E just in case there is at least one path from the C-variable to the E-variable, such that, given some fact about the values of off-path variables, E de facto depends on C, holding that fact fixed. They then state that it is clear according to this account that this proposal shows that in Figure 1, A is not a cause of E. (Figure 1 involves neuron E as dependent on D or B, B dependent on A but inhibited by C, and D dependent on C.) They say that no matter what fact we hold fixed about the off-path variables, E cannot depend on A. But this is not clear to me at all. If we hold that C and D do not fire, isn't it clear that E DOES depend on A? Because, holding that C and D do not fire, if B had not occurred then E would not have, and if A had not occurred then B would not have. Perhaps they mean that E depends on C intransitively in some way, whereas the dependence of E on A would have to be transitive (and hence problematic for other reasons). Is that what they mean? I don't know; they don't explain it. In cases like this it would be very helpful to have a little bit of explanation.
Additionally, there are several moments throughout the book where there are simply errors that seem to have just gotten through editing. Unfortunately, These errors weakened my trust in the authors. Whenever they said something I didn't immediately grasp, I had to check that some error of language wasn't at fault. Here are some examples.
In the List of Figures and on page 82, Figure 10 is mislabeled. It depicts neuron B as inhibitory and D as excitatory, despite its explanation in the List and its discussion in the text implying that it is the other way around. On further pages (198, 216, and 224) it is depicted the way the authors intended. This was very confusing to work out at the time.
On page 193 they say that on a minimum sufficiency account, D is not a cause of E's absence in Figure 22 because E would not have fired on such an account even without D's presence. But that's obviously true because D is not the inhibitory neuron; C is. They mean to be talking about C here, not D.
On page 207 they define an event, Victim's death, as V. But then they go on to talk about his death as event E, not event V. I am guessing that they initially labeled it V but then changed most of the instances of the use of that variable to E, but missed the one where they define it.
Most of these errors were not extremely frustrating (although the first one was), but were still not ideal. In any case, overall I think this is a good book on causation, and I have certainly learned a lot from reading it.
However, there are several problems with the book: an occasional lack of clarity and occasional mistakes. Regarding the first problem, sometimes Paul and Hall just did not explain themselves clearly. For instance, on pages 81-82 they state what they claim is Hitchcock's proposal of de facto causation: where C consists in some variable taking on a certain value, and E likewise consists of another variable taking on a certain value, C is a cause of E just in case there is at least one path from the C-variable to the E-variable, such that, given some fact about the values of off-path variables, E de facto depends on C, holding that fact fixed. They then state that it is clear according to this account that this proposal shows that in Figure 1, A is not a cause of E. (Figure 1 involves neuron E as dependent on D or B, B dependent on A but inhibited by C, and D dependent on C.) They say that no matter what fact we hold fixed about the off-path variables, E cannot depend on A. But this is not clear to me at all. If we hold that C and D do not fire, isn't it clear that E DOES depend on A? Because, holding that C and D do not fire, if B had not occurred then E would not have, and if A had not occurred then B would not have. Perhaps they mean that E depends on C intransitively in some way, whereas the dependence of E on A would have to be transitive (and hence problematic for other reasons). Is that what they mean? I don't know; they don't explain it. In cases like this it would be very helpful to have a little bit of explanation.
Additionally, there are several moments throughout the book where there are simply errors that seem to have just gotten through editing. Unfortunately, These errors weakened my trust in the authors. Whenever they said something I didn't immediately grasp, I had to check that some error of language wasn't at fault. Here are some examples.
In the List of Figures and on page 82, Figure 10 is mislabeled. It depicts neuron B as inhibitory and D as excitatory, despite its explanation in the List and its discussion in the text implying that it is the other way around. On further pages (198, 216, and 224) it is depicted the way the authors intended. This was very confusing to work out at the time.
On page 193 they say that on a minimum sufficiency account, D is not a cause of E's absence in Figure 22 because E would not have fired on such an account even without D's presence. But that's obviously true because D is not the inhibitory neuron; C is. They mean to be talking about C here, not D.
On page 207 they define an event, Victim's death, as V. But then they go on to talk about his death as event E, not event V. I am guessing that they initially labeled it V but then changed most of the instances of the use of that variable to E, but missed the one where they define it.
Most of these errors were not extremely frustrating (although the first one was), but were still not ideal. In any case, overall I think this is a good book on causation, and I have certainly learned a lot from reading it.
Want to read
September 11, 2020Need to take a step back. Pearl might be a good fit...
March 5, 2016
very good. wish they would've talked more about causation as applied to social sciences. need to be very fluent in metaphysics / first-order logic to read this book.
Displaying 1 - 3 of 3 reviews