IHMJ1007 The Philosophy of Mechanistic Explanation (1 cr)

The course is suitable for anyone who is interesting on “empirical” sciences or research.

It was once widely assumed that scientific theories are comprised of laws. The laws of nature were taken to be empirical and universal, i.e., unrestricted and exceptionless. The major account that explained how things actually happened with nomothetic settings (a world governed by laws) was Hempel and Oppenheim’s (1948) nomological-deductive (D-N) model. Popper (1934/1959) outlined similar ideas. For this reason, therefore, the deductive-nomological method of explanation is also sometimes called the Popper-Hempel theory of explanation.
In the 1960s (and later), the traditional view of theories as (primary) universal laws was challenged. Many of the best candidates for laws contain too many exceptions to be called bona fide laws of nature, or they apply in ideal rather than real settings. Especially, in life sciences, cognitive science, psychology and social science, the traditional concept of law have been found problematic. As a simple Information Systems example, consider ease of use, explaining IT use (from Technology Acceptance Model). This is not an exceptionless law in the sense of “all men are mortal”. What then is ease of use, explaining IT use? More generally, the question can be formulated as follows: what does the explanatory work in many sciences, if not laws? One answer is mechanism-based explanation (MBEs). MBEs, especially in the new mechanistic philosophy, are often distinguished from law-based explanations. Therefore MBEs should be interested especially for sciences that are aimed at explaining something without any laws of nature.

While I assume that the course is especially relevant for anyone doing empirical research that offer explanations, the course is interesting from the Information systems (IS) perspective for the following reasons. In IS models are often divided into variance and process models. MBEs seem to allow and, indeed, to explain alternative ways of modeling. Realizing this opens up new avenues for IS research, which could otherwise become unacceptable, because they do not meet existing IS conventions, such as laws (highlighted by Alan Hevner, Ron Weber, Arun Rai, etc), or variance models or process models. It is also worth emphasizing that MBEs contain typically, if not always, deliberate misrepresentations, called idealizations. In IS this observation is interesting, because idealizations (deliberate misrepresentations) seem to violate some fundamental tenets of existing IS philosophies.

In this short course, professor William Bechtel discusses what are mechanisms. Professor Bechtel proposed in 1980s that many explanations in sciences are not nomological, law-based explanations. Rather the scholars often explain by outlining the mechanisms, which describe how the phenomenon is produced. Our understanding of the MBEs has rapidly during the last 30 years. Professor Bechtel discusses the history of MBEs and some recent developments.

Content of the course:

PART I (AgC232.1)
1. Testing vs. Discovering Explanations
2. Laws, causes, and mechanisms
3. Investigating phenomena
4. Parts and operations: Decomposition as a discovery strategy
5. Organization and recomposition: diagrammatic representation
BREAK
6. Levels
7. Reduction and holism
8. Non-sequential organization and non-linear operations
9. Modeling mechanisms computationally: Dynamic mechanistic explanations
10. Possible vs. actual mechanisms
LUNCH
PART II (AgC233.1)
11. Energy and work: constraints
12. Mechanisms within autonomous systems (organisms)
13. Mechanisms within networks: network analysis
BREAK
14. Controlling mechanisms: Making measurements and working on flexible constraints
15. Organization of control systems: Hierarchy vs. heterarchy
16. Biological mechanisms and human-designed machines