Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions _quarto-website.yml
Original file line number Diff line number Diff line change
Expand Up @@ -14,6 +14,7 @@ project:
render:
- index.qmd
- intro-causal.qmd
- philosophy-of-causality.qmd
- collider-bias.qmd
- bias-analyses.qmd
- d_in_d.qmd
Expand All @@ -40,6 +41,8 @@ website:
menu:
- text: "Introduction to causal inference"
href: intro-causal.qmd
- text: "Philosophy of causality"
href: philosophy-of-causality.qmd
- text: "Collider bias"
href: collider-bias.qmd
- text: "Bias analyses"
Expand All @@ -55,6 +58,7 @@ website:
contents:
- index.qmd
- intro-causal.qmd
- philosophy-of-causality.qmd
- collider-bias.qmd
- bias-analyses.qmd
- d_in_d.qmd
Expand Down
11 changes: 11 additions & 0 deletions definitions/_def-counterfactual-account.qmd
Original file line number Diff line number Diff line change
@@ -0,0 +1,11 @@
:::{#def-counterfactual-account}

#### Counterfactual account of causation

On the **counterfactual account**,
event $A$ causes event $B$ if
$A$ and $B$ both occurred,
and, had $A$ not occurred,
$B$ would not have occurred.

:::
11 changes: 11 additions & 0 deletions definitions/_def-manipulability.qmd
Original file line number Diff line number Diff line change
@@ -0,0 +1,11 @@
:::{#def-manipulability}

#### Manipulability account of causation

On the **manipulability account**,
$A$ is a cause of $B$
if an intervention that changes $A$,
and that affects nothing else that bears on $B$,
would change $B$ [@woodward2003making].

:::
11 changes: 11 additions & 0 deletions definitions/_def-mechanistic-view.qmd
Original file line number Diff line number Diff line change
@@ -0,0 +1,11 @@
:::{#def-mechanistic-view}

#### Mechanistic view of causation

On the **mechanistic view**,
$A$ causes $B$ if
there is a mechanism,
an organized set of entities and activities,
that links $A$ to $B$ [@machamer2000thinking].

:::
10 changes: 10 additions & 0 deletions definitions/_def-probabilistic-view.qmd
Original file line number Diff line number Diff line change
@@ -0,0 +1,10 @@
:::{#def-probabilistic-view}

#### Probabilistic view of causation

On the **probabilistic view**,
$A$ causes $B$ if
$A$ raises the probability of $B$
after other relevant factors are held fixed [@suppes1970probabilistic].

:::
12 changes: 12 additions & 0 deletions definitions/_def-regularity-view.qmd
Original file line number Diff line number Diff line change
@@ -0,0 +1,12 @@
:::{#def-regularity-view}

#### Regularity view of causation

On the **regularity view**,
saying that $A$ causes $B$ means that
every event like $A$ is followed by an event like $B$
(a "constant conjunction"),
with $A$ coming first
and the two occurring close together in space and time.

:::
15 changes: 15 additions & 0 deletions definitions/_exm-counterfactual-account.qmd
Original file line number Diff line number Diff line change
@@ -0,0 +1,15 @@
:::{#exm-counterfactual-account}
<!-- no-figure: conceptual example with no data to plot -->

#### Counterfactual account and smoking cessation

Suppose a patient quit smoking
and did not develop lung cancer.
On the counterfactual account,
quitting prevented the cancer
only if the patient would have developed lung cancer
had they kept smoking.
We never observe that second scenario.
It is the [counterfactual outcome](intro-causal.qmd#def-counterfactual).

:::
19 changes: 19 additions & 0 deletions definitions/_exm-interventions.qmd
Original file line number Diff line number Diff line change
@@ -0,0 +1,19 @@
:::{#exm-interventions}
<!-- no-figure: conceptual example with no data to plot -->

#### A well-defined and an ill-defined question

"What would a patient's blood pressure be
if they took 10 mg of a particular antihypertensive daily for a year?"
names a specific intervention,
so $Y(a)$ is meaningful.

"What would a patient's blood pressure be
if they had a lower body weight?"
does not.
Diet, exercise, surgery, and illness all lower weight,
and they plausibly affect blood pressure differently.
We would need to specify a particular intervention
to define the causal question.

:::
15 changes: 15 additions & 0 deletions definitions/_exm-manipulability.qmd
Original file line number Diff line number Diff line change
@@ -0,0 +1,15 @@
:::{#exm-manipulability}
<!-- no-figure: conceptual example with no data to plot -->

#### Manipulability and a patient's sex

"What is the effect of a smoking-cessation program on lung cancer risk?"
names an intervention that can be assigned,
so it fits the manipulability account.

"What is the effect of a patient's sex on lung cancer risk?"
does not name an intervention.
On this account the question is not well posed
until we say which intervention is meant.

:::
14 changes: 14 additions & 0 deletions definitions/_exm-mechanistic-view.qmd
Original file line number Diff line number Diff line change
@@ -0,0 +1,14 @@
:::{#exm-mechanistic-view}
<!-- no-figure: conceptual example with no data to plot -->

#### Mechanism and smoking

A mechanistic account of smoking and lung cancer
would describe the steps between inhaling smoke
and the growth of a tumor,
such as damage to cells' DNA.
Evidence about such steps
can support the causal claim
alongside the statistical evidence.

:::
12 changes: 12 additions & 0 deletions definitions/_exm-probabilistic-view.qmd
Original file line number Diff line number Diff line change
@@ -0,0 +1,12 @@
:::{#exm-probabilistic-view}
<!-- no-figure: conceptual example with no data to plot -->

#### Probability raising and smoking

Smoking raises the probability of lung cancer
among people who are alike in age and other relevant factors,
and it does not make lung cancer certain.
Under the probabilistic view,
that is enough for smoking to be a cause of lung cancer.

:::
11 changes: 11 additions & 0 deletions definitions/_exm-regularity.qmd
Original file line number Diff line number Diff line change
@@ -0,0 +1,11 @@
:::{#exm-regularity}
<!-- no-figure: conceptual example with no data to plot -->

#### Regularity and smoking

Under a strict reading of the regularity view,
"smoking causes lung cancer" would be false,
because many smokers never develop lung cancer
and some non-smokers do.

:::
203 changes: 203 additions & 0 deletions philosophy-of-causality.qmd
Original file line number Diff line number Diff line change
@@ -0,0 +1,203 @@
---
title: "Philosophy of causality"
metadata-files:
- _chapter-formats.yml
format:
html: default
revealjs:
output-file: philosophy-of-causality-slides.html
pdf:
output-file: philosophy-of-causality-handout.pdf
---

{{< include latex-macros/macros.qmd >}}

::: {.callout-note}
## Optional reading

This chapter is background.
Nothing later in the book depends on it.
It explains where the potential-outcomes definitions in
[the introduction](intro-causal.qmd) come from,
and why they carry the assumptions they do.
:::

## Why look at philosophy at all?

The book defines a cause in terms of potential outcomes
([the definition of a cause](intro-causal.qmd#def-cause)).
That definition is one answer among several
to a question philosophers have argued over for centuries:
what does it mean to say that one thing causes another?
Seeing the other answers
makes it easier to see what the potential-outcomes answer includes,
what it leaves out,
and why.

Each account below is a family of views,
and this chapter simplifies all of them.
The cited sources give the full arguments.

## Five accounts of causation

### Regularity {#sec-regularity}

{{< include definitions/_def-regularity-view.qmd >}}

Hume's discussion of causation is the standard starting point
[@hume1748enquiry, sec. 7].
His point is that we never observe a necessary connection
between cause and effect.
We observe only that one kind of event reliably follows another.

{{< include definitions/_exm-regularity.qmd >}}

@exm-regularity shows the main weakness:
strict regularities are rare in epidemiology,
where outcomes are stochastic
and have many contributing causes.
Mackie's refinement treats a cause as
an insufficient but necessary part of a condition
that is itself unnecessary but sufficient for the effect
[@mackie1965causes].
That version fits "component causes" better.

### Counterfactual dependence {#sec-counterfactual-account}

{{< include definitions/_def-counterfactual-account.qmd >}}

Hume also wrote a second, counterfactual formulation of his definition
[@hume1748enquiry, sec. 7].
Lewis later developed it into a full theory
based on comparing the actual world
with the closest possible worlds in which $A$ does not occur
[@lewis1973causation].

{{< include definitions/_exm-counterfactual-account.qmd >}}

This account is the closest to [the definition of a cause](intro-causal.qmd#def-cause).
A potential outcome $Y(a')$ for an action $a'$ that did not occur
is a counterfactual outcome,
so the potential-outcomes framework makes counterfactual dependence
something a statistical model can describe
[@holland1986].

The account has two well-known difficulties.
First, "the closest possible world" is a philosophical notion
that has no obvious empirical meaning.
Second, it handles redundant causes badly:
if two sufficient causes were both present,
removing either one leaves the effect in place,
so neither counts as a cause.
The potential-outcomes framework avoids the first difficulty
by defining the counterfactual in terms of a specific action
([actions](intro-causal.qmd#def-action), @sec-manipulability),
and mostly sidesteps the second
by asking about the effect of one well-specified action at a time.

### Manipulation and intervention {#sec-manipulability}

{{< include definitions/_def-manipulability.qmd >}}

Here the cause is something that someone could in principle change.
Holland summarized this view as "no causation without manipulation"
[@holland1986].

{{< include definitions/_exm-manipulability.qmd >}}

This account is the one that the book's definitions
([actions](intro-causal.qmd#def-action) and [causes](intro-causal.qmd#def-cause)) take on:
a cause is an *action*,
and the effect is a contrast between outcomes under different actions.

### Probabilistic causation {#sec-probabilistic}

{{< include definitions/_def-probabilistic-view.qmd >}}

This view repairs the regularity view for stochastic outcomes.
It also matches how epidemiologists work:
a risk factor changes the probability of disease
and does not determine it.

{{< include definitions/_exm-probabilistic-view.qmd >}}

Its main difficulty is the one every observational study faces.
Probability raising can come from confounding,
so "after other relevant factors are held fixed" does the heavy lifting,
and the view does not say which factors those are.
[The chapter on bias analyses](bias-analyses.qmd)
discusses how to handle that question in practice.

### Mechanisms {#sec-mechanisms}

{{< include definitions/_def-mechanistic-view.qmd >}}

{{< include definitions/_exm-mechanistic-view.qmd >}}

Mechanistic evidence,
such as a known biological pathway,
often strengthens a causal conclusion
even when the statistical evidence is limited.
The potential-outcomes framework does not require a mechanism
to define an effect.
It defines the effect of an action on an outcome
without describing how the action works.

## What the potential-outcomes framework takes from these accounts

@tbl-accounts summarizes how each account relates to this book.

| Account | Core idea | Relation to this book |
|---|---|---|
| Regularity | Events of type $A$ are followed by events of type $B$ | Motivates asking whether an association is causal; too strict for stochastic outcomes |
| Counterfactual | $B$ would not have occurred without $A$ | Basis of [potential outcomes](intro-causal.qmd#def-potential-outcome) and [counterfactual outcomes](intro-causal.qmd#def-counterfactual) |
| Manipulability | A cause is something one could change | Basis of the definitions of [actions](intro-causal.qmd#def-action) and [causes](intro-causal.qmd#def-cause) |
| Probabilistic | Causes raise probabilities | Matches estimands defined on risks and means |
| Mechanistic | Causes work through mechanisms | Not needed for definitions; useful as supporting evidence |

: Philosophical accounts of causation and where they appear in this book. {#tbl-accounts}

Two assumptions in the book
come straight from the counterfactual and manipulability accounts:

- **Consistency** ([definition](intro-causal.qmd#def-consistency))
says that the outcome we observe under action $a$
equals the potential outcome $Y(a)$.
That statement is only meaningful
if "action $a$" names one thing.
@hernan2008obesity and @hernan2016water
argue this using examples such as obesity and body mass index:
different ways of changing the exposure
can lead to different outcomes,
so a question about "the" effect of the exposure is ambiguous
until the intervention is specified.
@vanderweele2013versions formalize the point as
the problem of multiple versions of treatment.
- **Well-defined interventions**
are the practical form of the manipulability account.
If we cannot say what intervention would set $A$ to $a$,
we cannot say what $Y(a)$ refers to,
and consistency has nothing to hold on to.

{{< include definitions/_exm-interventions.qmd >}}

## Not everyone agrees

The manipulability requirement is contested.
Some authors hold that causal questions about attributes
such as race or sex are meaningful without a specified intervention.
Others reply that
such questions can only mean something
once they are tied to an intervention on a related, manipulable variable.
This book follows the potential-outcomes tradition
and requires a specified intervention.
That choice is a design decision for the course,
and the other accounts above remain legitimate frameworks.

## Further reading

- The original discussion of the regularity and counterfactual formulations is Hume's *Enquiry* [@hume1748enquiry].
- @lewis1973causation is the standard statement of the counterfactual account.
- @woodward2003making develops the interventionist (manipulability) account in detail.
- @holland1986 connects the philosophical discussion to statistical causal inference.
- @hernan2016water gives an epidemiologist's argument for well-defined interventions.
Loading
Loading