What makes for a good scientific explanation?

A paper written for a philosophy of science course whilst studying for an MA at KCL. I have included it here because it picks up on many concerns which inform my approach to philosophy of art, in particular the epistemology of explanation and the
of 16
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
  What makes for a good scientic explanation? The basic task of an explanation is to give understanding for why a phenomenon occurs. We may know that   something is the case, but have no account of why this is so. For instance, I can form a belief that the sky is blue based on perceptual evidence or the testimony of others, but such knowing- that   may not allow me to explain why.  If we take answers to some form of why question to be explanations then many sorts of answer may be acceptable, including those based upon religious belief, aesthetic judgement, or legal reasoning. For philosophy of science the task is to lay out exactly how a why-question should best be answered, what makes such an answer specically scientic, and which account of explanation can best full these aims. As such, we are both concernedwhat explanations are, and the norms  of explanation.In this essay I will focus primarily on the latter concern. The importance of such atask has been clearly noted by William Wimsatt:An adequate philosophy of science should have normative force. It should help us to do science or, more likely, to nd and help us avoid sources of error, since scientic methodologies are by nature open-ended. Without being normative it is not a philosophical account. Mere descriptions of scientic practice, no matter how general or sensitive to detail, will not do.Without normative force, studies of methodology, however interesting, would translate as a catalogue of fortuitous and mysterious particular accidents, with no method at all. (Wimsatt, 2007: 26)I argue that the best way to account for ‘ good  scientic explanation’ is to consider how explanation can deal with the open-ended nature of scientic practice in a self-correcting and non-fortuitous manner. I take it that if something is open-ended it admits of much breadth and diversity methodology, and 1  dierent types of explanations. As such, my task is to see which theory of explanation can best formulate norms across this variation. We must also note that one’s stance on explanation entails diering epistemological and ontological commitments. Explanations can serve an epistemological purpose insofar as they are used, via models, representations, orother activities, to communicate an understanding of a particular phenomenon.  They serve an ontological purpose insofar as they explain an aspect or structure of the world which is responsible for a worldly event. In this essay I will argue that developing the norms of scientic explanation is best served by an account which adequately integrate the heuristics of scientic practice into explanation, and argue against accounts which take only ontological concerns to be primary. This essay will rst focus on the relation between two strategies of explanation put forward by philosophers of science: the deductive-nomological (D-N) model of explanation, and mechanistic explanation. I will argue that the latter provides a better ability to account for the diversity of   scientic practice, yielding less paradoxes than the former. Second, I will focus on the internal debates within mechanism concerning the relation between heuristics and ontology. I will argue that the close link between the mechanisms and the investigation of the nature of the human rationality and heuristics involved in forming explanations yields a reexive picture of scientic practice which is best suited to understanding its normative status. 1.1 The Deductive-Nomological Model 2  In their 1948 paper ‘Studies in the Logic of Scientic Explanation’ Carl G. Hempeland Paul Oppenheim attempted to integrate explanation into a logical empiricist picture of science with their deductive-nomological theory of explanation. On thisview, a good explanation will consist of two things. First, it will be a valid, deductive logical argument in which explanans  (the things that do the explaining, which must include some empirical content) will constitute the premises, and the explanandum  (the thing to be explained) will be the conclusion. Secondly, the premises must contain at least one statement of a law of nature. We may conceive of a law of nature as a regularity or pattern in the ow of events as registered by scientic practice, which can support counterfactual inferences and modal statements of physical necessity and impossibility.Consider this example of D-N explanation. We observe a gure skater balanced on one skate. They begin spinning by pushing themselves with the other skate.  They stop pushing and continue to spin slowly. They then pull their arms in close and beginning spinning faster. Why do they speed up? A basic D-N explanation would go as follows:1.The angular momentum of any body remains constant (unless there are external forces acting upon it).2.The skater is not interacting with any external object in such a way as to alter their angular velocity.3.The skater is rotating (her angular momentum is non-zero).4.The skater reduces their moment of inertia by drawing their arms close to their body5.It follows from 1-4 that skater’s rate of rotation increases. 3  Here premise 1 states the law of the conservation of angular motion and premises 2, 3, and 4 supply the necessary empirical data and their consequences in regards to the law. From this we condently derive our conclusion. Even in this simplied form, one can see that the D-N model has the virtues of being clear and intuitive. It can also accommodate and account for the predictivecapacities of scientic practice (by its lights “explanation simply is a form of prediction.” (Halina, 2018: 216)). Furthermore, it ts well with statistical explanations used across scientic practice since we can simply replace the natural law with a statistical law and take the result to be an inductive inference and, as such, we can apply the D-N model to a number of cases from various scientic disciplines like medicine where probability may account for explanations of a proposed treatment’s success, or evolutionary biology where probability can be used to investigate a variety of phenomena, such as an individual of a population obtaining a particular mutation. 1 1.2. Counter Examples However, not long after Hempel and Oppenheim canvassed the D-N theory counterexamples began to emerge which fullled all the conditions for D-N explanation, but failed to qualify as explanations. For instance, if asked why the shadow of a certain agpole was the length it was we could construct the following valid D-N explanation: 1  This latter example is in reference to Durrett and Schmidt (2008) 4  1.On a at, level piece of ground stands a 12’ ag pole.2.The sun is at an elevation of 53.13º in the sky, shining brightly.3.The shadow is 9’ long.4.Given the laws of trigonometry, and premises 1, 2, and 3, we can deduce that length of the shadow is 9’ long.But we can ask a dierent questions about the same situation: why is the agpole 12’? Why is the sun at an elevation of 53.13º? On the D-N model we can give exactly the same valid argument as above, switching the objects where appropriate, to explain both these phenomena. It would be doubtful that many would the length of a particular shadow as good explanation of the agpole’s height or the sun’s elevation. What has gone wrong? In the asking about the shadow it is clear that we are not just explaining the nature of the phenomenon, but also pointing to causes for it. In the other two cases it would be hard to take them to be eects of the length of the shadow. 1.3. Grounds for rejecting the D-N model  The D-N model is thus not the best account of good  scientic explanation. Whilst it can lead to many valid explanations and can appreciate the statistical methodsused across the sciences, it can also be found by its own lights to produce valid arguments that do no explanatory work or, worse, create misleading explanations. We have yielded only a very partial picture of explanation, and though we can grasp how we may derive scientic explanations we have no means for saying that they are good explanations. Moreover, a problem with this model is that whilst it seems it can account for wide range of scientic practices, it can only really account for those which do indeed allow us to reduce our 5
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!