Crash Course: Causation

Crash Course: Causation

A few weeks ago we started a new series of “crash course” posts here at Daily Nous. The idea is borrowed from Natalia Cecire (Sussex): to come up with a “one-week self education program” for “students who suddenly need to get up to speed in a field, and don’t have time to take a course or immerse themselves in it for a year,” or for professors seeking to learn about a new area for teaching or research.

Putting together the crash course is a kind of challenge. Some characteristics of the “crash course”:

  • it should contain substantive primary works on the subject
  • it should be reasonable to expect someone to complete the set of readings in about a week
  • it should not include introductory texts (not because they’re not useful, but because, first, everyone knows to look for them, and second, they are boring as suggestions)

Cecire’s sample syllabi (links here) also include a small set of questions central to the subdisciplinary area that the student in the crash course should keep in mind while doing the readings.

This week let’s build a crash course on causation. Your suggestions, please.

(image: photo of book dominoes at the Seattle Public Library, from Huffington Post)

Newest Most Voted
Inline Feedbacks
View all comments
5 years ago

If this crash course is intended for a general readership, I suggest it would be catch readers’ interest most effectively if each section began by setting out a real practical problem or controversy taken from science or law concerning finding the cause of something, or attributing legal culpability, or finding an explanation for a phenomenon, or finding the reasons for a correlation, or predicting events or controlling events or constructing an argument. The problem-based approach would immediately focus the reader’s attention on the reasons why he needs to know about causality, and what things he needs to know about it. The didactic bits could come out in the process of struggling with the practical problems.Report

Justin philgradcandidate
Justin philgradcandidate
5 years ago

Excellent idea, Ian!Report

Roberta L Millstein
5 years ago

Back in 2009, I did a grad seminar on causation with the following readings. Here are the readings I used, with the caveat that I might do it differently today:

Dependence causes: Ronald Giere, “Causal Systems and Statistical Hypotheses” and David Lewis, “Causation as Influence”
Dependence causes: Jim Woodward, “Causation with a Human Face” and Kenneth Reisman & Patrick Forber, “Manipulation and the Causes of Evolution”
Production causes: Wesley Salmon, “Causality without Counterfactuals” and Phil Dowe, “Causality and Conserved Quantities: A Reply to Salmon”
Production causes: Stuart Glennan, “Mechanisms and the Nature of Causation” and Peter Machamer, Lindley Darden, & Carl Craver, “Thinking about Mechanisms”
Probabilistic causation: Elliott Sober, “Two Concepts of Cause” and Christopher R Hitchcock, “A Generalized Probabilistic Theory of Causal Relevance”
Probabilistic causation: Peter Menzies, “Probabilistic Causation and the Pre-emption Problem” and Ellery Eells, “Propensity, Preemption, and the Identity of Events”
Applications: M Parascandola and DL Weed, “Causation in Epidemiology” and Daniel Steel, “Social Mechanisms and Causal Inference”
Pluralism: Ned Hall, “Two Concepts of Causation” and Nancy Cartwright, “Causation: One Word, Many Things”
Anti-pluralism: Stathis Psillos, “A Glimpse of the Secret Connexion: Harmonizing Mechanisms with Counterfactuals” and John Gerring, “Causation: A Unified Framework for the Social Sciences”Report

Roberta L Millstein
5 years ago

Then, in 2011, I did a seminar focused on mechanisms (same caveat as above):

Darden 2008 Thinking again about biological mechanisms
Hitchcock 1995 Salmon on Explanatory Relevance
Machamer, Darden, & Craver 2000 Thinking about mechanisms
Glennan 1996 Mechanisms and the Nature of Causation
Fehr 2004 Feminism and Science: Mechanism Without Reductionism
Bechtel and Abrahamsen 2005 Explanation: a mechanist alternative
Craver 2007, Ch. 3 Causal Relevance and Manipulation
Craver 2007, Ch. 4 The Norms of Mechanistic Explanation
Craver 2007, Ch. 5 A Field Guide to Levels
Craver and Bechtel 2007 Top-down causation without top-down causes
Steel 2008, Ch. 1 Extrapolation and Heterogeneity (short intro to book)
Steel 2008, Ch. 2 Interventions, Causal Effects, and Causal Relevance
Steel 2008, Ch. 3 Causal Structure and Mechanisms
Steel 2008, Ch. 4 The Disruption Principle
Steel 2008, Ch. 5 Extrapolation, Capacities, and Mechanisms
Skipper and Millstein 2005 Thinking about evolutionary mechanisms: natural selection
Glennan 2009 Productivity, relevance and natural selection
Illari and Williamson 2010 Function and organization: comparing the mechanisms of protein synthesis and natural selection
Craver 2009 Mechanisms and natural kinds
Torres 2009 A Modified Conception of Mechanisms
Dowe 2010 The causal-process-model theory of mechanismsReport

Jonathan Livengood
Jonathan Livengood
5 years ago

I don’t think an adequate crash course on causation can be done. But here are some thoughts. The following should be supplemented by reading the Oxford Handbook.

On Hume:
Robinson (1962) “Hume’s Two Definitions of ‘Cause'” Phil Quarterly 12, 162-171.
Beebee (2007) “Hume on Causation,” Chapter 9 in Price and Corry (eds) Causation, Physics, and the Constitution of Reality.

On Regularity Theories:
Davidson (1967) “Causal Relations” J Phil 64(21), 691-703.
Baumgartner (2008) “Regularity Theories Reassessed” Philosophia 36, 327-354.

On Probabilistic Theories:
Salmon (1980) “Probabilistic Causality,” Pacific Phil Quarterly 61, 50-74.
[[Lots of other stuff, but in my opinion, reductive probabilistic approaches are dead, having been superseded by non-reductive approaches in the graphical modeling tradition.]]

On Counterfactual Theories:
Lewis (1973) “Causation,” J Phil 70, 556-567.
Lewis (2000) “Causation as Influence,” J Phil 94(4), 182-197.
Beebee (2006) “Does anything hold the universe together?” Synthese 149, 509-533.

On Agency and Interventionist Theories:
Ducasse (1926) “On the Nature and Observability of the Causal Relation,” J Phil 23, 57-68.
Cartwright (1979) “Causal laws and effective strategies,” Nous 13, 419-437.
Hausman (1986) “Causation and experimentation,” Am Phil Quarterly 23, 143-154.
Price (1991) “Agency and Probabilistic Causality,” BJPS 42, 157-176.
Meek and Glymour (1994) “Conditioning and Intervening,” BJPS 45, 1001-1021.

On Process and Mechanism Theories:
Salmon (1994) “Causality without counterfactuals,” Phil Sci 61, 297-312.
Machamer et al. (2000) “Thinking about mechanisms,” Phil Sci 67, 1-25.
Woodward (2002) “What is a mechanism?” Phil Sci 69, S366-S377.
Psillos (2004) “A glimpse of the secret connection,” Perspectives on Science 12, 288-319.
[[A bunch of stuff by Dowe … hard to choose.]]

On Causal Powers and Capacities:
Harre (1970) “Powers,” BJPS 21, 81-101.
Lewis (1997) “Finkish Dispositions,” Phil Quarterly 47, 143-158.
Hiddleston (2005) “Causal Powers,” BJPS 56, 27-59.

On Anti-Reductionism, Pluralism, and Skepticism:
Anscombe (1971) “Causality and Determination,” in Sosa and Tooley (eds) Causation.
Tooley (1990) “Causation: Reduction versus Realism,” PPR 50, 215-236.
Hitchcock (2003) “Of Humean Bondage,” BJPS 54, 1-25.
Cartwright (2004) “One Word, Many Things,” Phil Sci 71, 805-819.
Russell (1912) “On the Notion of Cause,” Proceedings of the Aristotelian Society 13, 1-26.
Skyrms (1984) “EPR: Lessons for Metaphysics,” Midwest Studies 9, 245-255.
Norton (2003) “Causation as Folk Science,” Phil Imprint 3. [Reprinted as Chapter 2 of Price and Corry]
Hitchcock (2007) “What Russell Got Right,” Chapter 3 in Price and Corry.

On Causal Modeling and Statistics:
Holland (1986) “Statistics and Causal Inference,” J of the Am Stat Assoc 81, 945-960.
Freedman (1991) “Statistical Models and Shoe Leather,” Sociological Methodology 21, 291-313.
Pearl and Verma (1991) “A Theory of Inferred Causation” [[available online]]
Pearl (1995) “Causal Diagrams for Empirical Research,” Biometrika 82, 669-688.
Dawid (2000) “Causal Inference without Counterfactuals,” J of Am Stat Assoc 95, 407-424.
Pearl (2001) “Bayesianism and Causality,” in Foundations of Bayesianism
Scheines (2002) “Computation and Causation,” Metaphilosophy 33, 158-180.
Spirtes (2010) “Introduction to Causal Inference,” J of Machine Learning Research 11, 1643-1662.
Greenland (2011) “The Logic and Philosophy of Causal Inference,” Chapter 25 in Bandyopadhyay and Forster (eds) Philosophy of Statistics
[[Lots and lots of great, even more technical stuff by Glymour, Spirtes, Scheines, Eberhardt, Richardson, Zhang, and others. Too much to list, really.]]

On Actual Causation:
Halpern and Pearl (2005) “Causes and Explanations: A Structural-Model Approach, Parts 1 and 2,” BJPS 56, 843-887 and 889-911.
Hitchcock (2007) “Prevention, Preemption, and the Principle of Sufficient Reason,” Phil Review 116, 495-532.
Hall (2007) “Structural Equations and Causation,” Phil Studies 132, 109-136.
Halpern (2008) “Defaults and Normality in Causal Structures,” Proceedings of the 11th International Conference on Principles of Knowledge Representation and Reasoning.
Hitchcock and Knobe (2009) “Cause and Norm,” J Phil 61, 587-612.
Glymour et al. (2010) “Actual Causation: A Stone Soup Essay,” Synthese 175, 169-192.
Halpern and Hitchcock (2014) “Graded Causation and Defaults” BJPS axt.
Blanchard and Schaffer (forthcoming?) “Cause without Default”
Halpern (2014?) “Appropriate Causal Models and the Stability of Causation”

On Causation and Responsibility:
Moore (1999) “Causation and Responsibility,” Social Philosophy and Policy 16, 1-51.
Sartorio (2004) “How to be responsible for something without causing it,” Phil Perspectives 18, 315-336.
Chockler and Halpern (2004) “Responsibility and Blame,” J of Artificial Intelligence Research 22, 93-115.
Sartorio (2007) “Causation and Responsibility,” Phil Compass 2/5, 749-765.
[[There is so much more here, with tons of overlap with the second section on psychology, it’s just not at all clear how to start.]]

On Causation and Psychology (Learning):
Scholl and Tremoulet (2000) “Perceptual Causality and Animacy,” Trends in Cog Sci 4, 299-309.
Gopnik et al. (2004) “A Theory of Causal Learning in Children,” Psych Review 111, 3-32.
Penn and Povinelli (2007) “Causal Cognition in Humans and Nonhuman Animals,” Annual Review of Psych 58, 97-118.

On Causation and Psychology (Attribution):
Livengood and Rose (forthcoming) “Experimental Philosophy and Causal Attribution” in Sytsma and Buckwalter (eds) Companion to Experimental Philosophy
Kelley (1973) “The Processes of Causal Attribution,” Am Psychologist 28, 107.
Einhorn and Hogarth (1986) “Judging Probable Cause,” Psych Bulletin 99, 3-19.
Hilton and Slugoski (1986) “Knowledge-Based Causal Attribution,” Psych Review 93, 75-88.
Alicke (1992) “Culpable Causation,” J of Personality and Soc Psych 63, 368-378.
Ahn et al. (1995) “The Role of Covariation Versus Mechanism Information in Causal Attribution,” Cognition 54, 299-352.
Gerstenberg and Lagnado (2010) “Spreading the Blame,” Cognition 115, 166-171.
Danks et al. (2014) “Demoralizing Causation,” Phil Studies 171, 251-277.

On Meta-Theory for Causation:
Glymour (2004) “Review of Woodward,” BJPS 55, 779-790.
Hall (2004) “Rescued from the Rubbish Bin,” Phil Sci 71, 1107-1114.
Menzies chapter (“Platitudes and Counterexamples”) from Oxford Handbook
Paul (2010) “A New Role for Experimental Work in Metaphysics,” Review of Philosophy and Psychology 1, 461-476.
Paul (2012) “Metaphysics as Modeling,” Phil Studies 160, 1-29.
Chapters 1 and 2 from Hall and Paul (2013) Causation: A User’s Guide

This list leaves out SO MUCH. I feel embarrassed offering it at all. Lots of stuff on causation and explanation, on causation by absence, omission, and disconnection, on backwards causation, on the causal Markov condition (and related assumptions for causal inference), and on and on and on. This is part of why I think a crash course on “causation” without further qualification just can’t be done.Report

Grad Sockpuppet
Grad Sockpuppet
5 years ago

I have a really hard time understanding why we seem to be having so much trouble with these; it’s like we’re incapable of understanding the concept. These “crash courses” aren’t meant to be comprehensive or definitive or year-long. They’re just meant to be fairly advanced introductions to a topic, so that advanced students (honours thesis-writing undergrads, MA students) can quickly (viz. in one week of reading) gain some more purchase on a field than they would by merely reading the Wikipedia or SEP entry. The key here is that the crash courses are meant to help people gain some purchase in an area in a short period of time (***one week***). Maybe it’s hard to agree on what would be good, or even on which things are necessary, but come on! It’s not *that* hard to follow the instructions!

Look at the examples. They all have the same format: six readings, four discussion-style questions to which written responses are suggested (to get the student thinking along the right/important lines, especially to guide further inquiry), a directive to glance at the contents of two major journals in the field in question, and then instructions for building a list of further reading. It’s not *that* hard. Geeze.

If I knew anything at all about causation, I’d lead by example. I don’t, but here’s a list of six readings distilled from the above to get you started properly:

1. Robinson – Hume’s Two Definitions of ‘Cause’
2. Davidson – Causal Relations
3. Salmon – Probabilistic Causality
4. Lewis – Causation
5. Cartwright – Causal laws and effective strategies
6. Salmon – Causality without counterfactuals

Boom. Done. You can put together the other sections yourselves, and use them to point students towards questions raised in the other readings you all suggested.

(I’m sorry about my tone, but I’m finding these threads really, really exasperating. Just follow the bloody instructions, already–or at least attempt to!)Report

Jonathan Livengood
Jonathan Livengood
5 years ago

Grad Sockpuppet,

I’m sorry that you’re exasperated, but I don’t think your suggested list would work. If an instructor used just those six readings, the result would be a very, very odd view of causation research. You seem to have constructed your list by taking one reading from each of the first six sub-headings in my long list. How do you propose to get students from those to readings falling under the other eight sub-headings? Or more particularly, how do you do that without just reproducing the long list (or something similar)?

My point in offering my list was to support the claim that a crash course that is both doable in a week and also suggests even the basic geography of the topic is impossible for causation. It should not be surprising, therefore, that I didn’t abide by the rules for describing a crash course for causation.

I think it might be possible to give a crash course for some sub-topic in causation. You might read my list as giving the beginnings of 14 separate crash courses. (Guiding questions would have to be added, and in some cases, choices would have to be made to narrow even the sub-lists. And I already felt like I had done a TON of pruning to get those sub-lists to be as short as they are.) But there just isn’t going to be a workable crash course on causation as a whole. Any six readings that anyone might name will fail to even hint at most of the *kinds* of literature, or even mention most of the interesting literature in bibliographies. The result of trying to do a crash course on a too-large topic will be to distort what the literature looks like and/or to give a very biased view of what is important or interesting. I suspect that the large list I gave is itself massively idiosyncratic and leaves out important literature. (There are no pieces by Yablo or Schaffer; there are no pieces on specific problems like mental causation; there are no pieces on more abstract questions of causal metaphysics, like what sorts of things could be the relata of a causal relation; and there are almost certainly other things left off the list because I haven’t even considered them.)Report

Grad Sockpuppet
Grad Sockpuppet
5 years ago


It seems to me that the same is going to be true of pretty much any other topic we choose unless it’s too narrow to fulfill the desiderata of the crash course project. That’s probably also true of the original crash courses (actually, the philosophy topics so far are already much narrower than those). The topic of causation is no exception; six just isn’t a lot of readings, even when supplemented by the rest of the work suggested. But I don’t think that means it’s not a worthwhile exercise, and worth taking up seriously.

I guess I think you’re making too much of the goals here. (Or maybe I’m making too little of them…) And that’s what’s frustrating me about these threads: people are just giving a lengthy laundry list of readings without much regard for the purpose of the exercise. If one needs 10 to get the job done rather than 6, that seems a bit much (given the rest of the exercises) but also OK (hell, we could just include the SEP entry, even though that should be someone’s first stop anyway. Surely that manages to paint an adequate preliminary picture?), but 71 is just unreasonable. At that point, we’re outlining a reading list for a dissertation on the topic. And that, in turn, makes it harder to narrow the list down.

I mean, suppose you had to give a full-semester course on causation. You couldn’t hope to assign all the work you suggested, but presumably you could narrow it down to a semester’s worth of reading. Why not start with that, and then see what it would look like if it was halved? That’s still more than we want, but it’s a whole lot closer to the goal!Report