How mechanisms answer questions Matej Kohar Models of Explanation, - - PowerPoint PPT Presentation

How mechanisms answer questions

Matej Kohar Models of Explanation, T urin 12th June 2018

Problem

Why X?

• Because Y.

?? ??

2

Outline

• 1. Mechanistic explanation
• 2. Pragmatics of explanatory speech-

acts

• 3. Wrong solutions
• 4. The solution

3

• 1. Mechanistic

explanation

“A mechanism is a structure performing a function in virtue of its component parts, component operations, and their organization. The orchestrated functioning of the mechanism is responsible for one or more phenomena.” (Bechtel and Abrahamsen 2005: 423) “In biology, identifying phenomena precedes and invites explaining them. … [T]he term biologists most frequently invoke in explanatory contexts is mechanism. … [B]iologists explain why by explaining how.” (ibid.: 422)

4

• 1. Mechanistic

explanation

• Explanandum phenomenon
• Explanans mechanism
• Causal/constitutive relation
• Entities and activities appropriately
• rganised

5

• 2. Pragmatics of explanatory

speech acts (van Fraassen 1977, 1980)

• Explanatory requests: <A, X, R>

– A: the topic (explanandum) – X: contrast class – R: relevance relation

6

• 2. Pragmatics of explanatory

speech acts

• Answer: “Because B.”

– A is the case – ⱯxX: x → x = A – R(B, A)

• Note: not good answer, just an answer
• Evaluation criteria available, but R

eliminates most bad answers

• R determines which evaluation criteria

to use

7

• 4. Solution
• Basic idea: identify a set of changes to the

mechanism for the phenomenon, which, had they been actual, would change it into a phenomenon that belongs to the contrast class

• Afnity with mutual manipulability

strategy for discovery – many explanations are found in top-down experiments

15

• 4. Solution
• More precise characterisation needed
• Slight diferences between

explanation of token phenomena and explanation of regularities require separate treatment

16

4.1 T

• ken phenomena
• Explanatory request: <f, G>

f: a token phenomenon G: a contrast class of phenomena defned by membership conditions

• f ∉ G
• G can be empty
• Why did f occur rather than some g ∊

G?

17

4.1 T

• ken phenomena
• Answer: “Because Cf rather than CG

Cf: a subset of components of mechanism Mf for phenomenon f CG: a set of counterfactual mechanism components

• If Cf were replaced by CG in Mf, the

resulting mechanism would underlie a phenomenon g ∊ G

18

4.2 Regularities

• Diferences:

– Multiple question types – Mechanistic models, instead of mechanisms as such are involved in explanation

• Explaining a regularity: answering an

explanation request about multiple token phenomena

19

4.2 Regularities - questions

• T

ypes of explanatory request about regularities:

– substantive – frequency – timing

20

4.2 Regularities - questions

• Substantive explanatory request:

<F, G>

F: a class of explanandum phenomena defned by membership conditions G: a contrast class of phenomena defned by membership conditions

• F ∩ G = ∅
• G can be empty

21

4.2 Regularities - questions

• Frequency explanatory request:

<F, φ, φ*>

F: A class of explanandum phenomena defned by membership conditions φ: the frequency at which instances of F

• ccur

φ*: a contrast frequency φ* ≠ φ

• φ* can be 0

22

4.2 Regularities - questions

Timing explanatory request: <F, T, T*>

F: a class of explanandum phenomena defned by membership conditions T: a set of time intervals during which instances of F occur T*: a contrast set of time intervals

• T* ≠ T

23

4.2 Regularities - answers

Substantive explanatory request: “Because CF rather than CG” CF: a subset of components of MF – the mechanistic model for class F CG: a set of components such that any phenomenon underlain by a mechanism adequately represented by the mechanistic model MG = MF – CF + CG would be a member of class G.

24

4.2 Regularities - answers

Frequency explanatory request: “Because set-up conditions for mechanisms adequately represented by MF, the mechanistic model of F, occur with frequency φ rather than φ*”

• Can be supplemented by etiological

mechanistic explanation of the set-up conditions

• Can be treated as an instance of explaining

token phenomena

25

4.2 Regularities - answers

Timing explanatory request: “Because set-up and termination conditions for mechanisms adequately represented by MF, the mechanistic model of class F, specify intervals in T rather than intervals in T*”

• Can be flled in by an account of how conditions

in T* would have to change to trigger mechanisms adequately represented by MF,

• or how MF would have to change in order for

mechanisms adequately represented by it to be triggered in T*

26

4.2 Regularities – mechanistic models

• See Glennan (2017)
• Obtained from token mechanisms by

abstraction

• Properties of components specifed in

a less determinate way, so that the description which specifes the mechanistic model is similar to actual members of the modelled class

27

Thanks for attention!

References

 Bechtel, W. & Abrahamsen, A. (2005). Explanation: a

mechanist alternative. Studies in History and Philosophy

• f Biology and Biological Sciences 36(2), 421-41.

 Glennan, S. (2017). The New Mechanical Philosophy.

Oxford: Oxford University Press.

 Van Fraassen, B. (1977). Pragmatics of explanation.

American Philosophical Quarterly 14(2), 143-50

 Van Fraassen, B. (1980). The Scientific Image. Oxford:

Clarendon.