Computer Simulations in Science and Engineering by Juan Manuel Durán

Author:Juan Manuel Durán
Language: eng
Format: epub
ISBN: 9783319908823
Publisher: Springer International Publishing

4.3.2 Epistemic Opacity

The previous discussion was an effort to show how errors can contribute to the general inaccuracy of results and thus compromise the reliability of computer simulations. As presented, there are as many sources of errors as there are ways of dealing with them. All things considered, we can justifiably say that many errors—but not all, naturally—are corrigible to a given extent, and therefore they are not so critical to the reliability of computer simulations. Unfortunately, in computational science—and therefore in computer simulations—there is a far more worrying source of mistrust than errors, that is, epistemic opacity.

The history of this concept goes back well before the use of computers for scientific purposes. However, it was Paul Humphreys who introduced the term as a distinctive mark of computer science (Humphreys 2004). According to him, an essential feature of epistemic opacity is that researchers are unable, as cognitively limited human beings, to know all the relevant states of a given computational process at any given time. The argument is quite compelling. It says that no human—or group of humans—could possibly examine every element of the computational process relevant for the assessment and justification of the results. Again, ‘justification of the results’ here simply means to have reasons to believe that the results are correct. Epistemic opacity, then, is conceived as the unrecoverable loss of knowledge about a given computational process followed by the inability to justify the results of such a process (Humphreys 2004, 148). To characterize epistemic opacity more formally, I reproduce Humphrey’s definition:a process is essentially epistemically opaque to [a cognitive agent] X if and only if it is impossible, given the nature of X, for X to know all of the epistemically relevant elements of the process. (Humphreys 2009, 618)

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.