Causation in Science – On the Methods of Scientific Discovery

A book by Stephen Mumford and Rani Lill Anjum, forthcoming at Oxford University Press, 2017

Some of the chief goals of science are understanding, explanation, prediction and application in new technologies. Only if the world has some significant degree of constancy in what follows from what can these scientific activities be conducted with any purpose. But what is the source of such predictability and how does it operate? In many ways, this is a question that goes beyond science itself – beyond the data – and inevitably requires a philosophical approach. This book argues in such terms that causation is the main foundation upon which science is based.

I. SCIENCE AND PHILOSOPHY

1.       Metascience and better science PHILOSOPHY – SCIENCE – REFLECTIVE EQUILIBRIUM

2.       Do we need causation in science? OBSERVATION – EMPIRICISM – SCEPTICISM

3.       Evidence of causation is not causation ONTOLOGY – EPISTEMOLOGY – DISCOVERY

 

II. PERFECT CORRELATION

4.       What’s in a correlation? STATISTICS – COINCIDENCES – HILL’S CRITERIA

5.       Same cause, same effect CONSTANCY – CONJUNCTION – NOISE

6.       Under ideal conditions IDEALISATIONS – CLOSED SYSTEM – EXCEPTIONS

7.       One cause, one effect? It’s complicated COMPLEXITY – TOTAL CAUSE – EXPLANATION

 

III. INTERVENTION AND PREVENTION

8.       How to have your cause and beat it INTERFERENCE – CONTEXT-SENSITIVITY – NONLINEARITY

9.       Correlation is not causation CONSTANT CONJUNCTION – INCIDENCE – TENDENCIES

10.   What to look for when searching for causation? DISPOSITIONALITY – NECESSITY – POWERS

 

IV. CAUSAL MECHANISMS

11.   Is the business of science to construct theories? RAW DATA – THEORY-CONSTRUCTION – PREDICTION

12.   How much data do we need? HUMEANISM – SINGULARISM – ONTOLOGY

13.   The explanatory power of mechanisms MECHANISM – STATISTICS – EVIDENCE

14.   Digging deeper to find the real causes REDUCTIONISM – HOLISM – EMERGENCE – LEVELS

 

V. LINKING THE CAUSE TO ITS EFFECT

15.   Do causes make a difference? RCT – COUNTERFACTUALS – OVERDETERMINATION

16.   Making nothing happen EQUILIBRIUM – HOMEOSTASIS – STABILITY

17.   It all started with a Big Bang CAUSAL CHAINS – DETERMINISM – TRANSITIVITY

18.   Does science need laws of nature?  SINGULARISM – PROPERTIES – COVERING LAW

 

VI. PROBABILITY

19.   Probably true or probably right? PROBABILITY – CHANCE – CREDENCE

20.   Where to look for probabilistic causation STATISTICS – FREQUENTISM – PROPENSITIES

21.   Calculating conditional probability is no simple matter CONDITIONAL PROBABILITY – RATIO – ONTOLOGY

 

VII. EXTERNAL VALIDITY

22.   Risky predictions EXTERNAL VALIDITY – UNCERTAINTY – CLOSED SYSTEMS

23.   What RCTs do not show HETEROGENEITY – COMPLEXITY – MARGINALS – N OF 1

 

VIII. DISCOVERING CAUSES

24.   Uncovering causal powers INTERVENTION – INVENTION – HIDDEN POWERS – SIDE EFFECTS

25.   Getting involved PROCESSED DATA – PRESUPPOSITIONS – MANIPULATION

26.   Different methods, same evidence? PLURALISM – CONFLICTING EVIDENCE – METHODS

27.   When causation fails NEGATIVE RESULTS – THEORY DEVELOPMENT – PROGRESS

28.   Understanding causation by way of failure  FAILED EXPERIMENT – DISCOVERY – REPRODUCIBILITY

 

This book will be the culmination of the research carried out on the NFR-funded Causation in Science (CauSci) project, a 4-year interdisciplinary project.

 

http://www.forskningsradet.no/en/Home_page/1177315753906
http://www.nmbu.no/causci