Abstract
One staple of living in our information society is having access to the web. Web-connected devices interpret our queries and retrieve information from the web in response. Today’s web devices even purport to answer our queries directly without requiring us to comb through search results in order to find the information we want. How do we know whether a web device is trustworthy? One way to know is to learn why the device is trustworthy by inspecting its inner workings (Lehrer The Monist, 78(2), 156–170 1995; Humphreys 2004, Episteme, 6(2), 221–229 2009). But ordinary users of web devices cannot inspect their inner workings because of their scale, complexity, and the corporate secrecy which enshrouds both the procedures by which the devices operate and the companies that make them (Pasquale 2015). Further piling on this predicament, authors have criticized web technology on the grounds that the invisibility of the web devices’ inner workings prevents users from critically assessing the procedures that produce a given output, in some cases, barring users from fulfilling their epistemic responsibilities (Simon Ethics and Information Technology, 12(4), 343–355 2010; Miller and Record Episteme, 10(2), 117–134 2013). I consider four broad kinds of reasons which we can acquire without inspecting the inner workings of black-boxed technology: individual understanding, expert testimony, testing through experience, and social vetting; and show how each is a viable method of appraising black-boxed technology. By deploying these methods, we can remain responsible inquirers while nonetheless benefitting from today’s epistemic resources on the web.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Explore related subjects
Discover the latest articles and news from researchers in related subjects, suggested using machine learning.Notes
Here I have in mind specifically Google Search and its related devices. For Google’s own account of how search works, see https://www.google.com/intl/bn/insidesearch/, see the slogan for the Google app: “In the latest version of the Google app for iOS and Android, not only can you say your question out loud, but your search app can speak your answer right back to you. And, using Google’s Knowledge Graph, your search app gives you smarter answers loud and clear.” Accessed December 14, 2016.
I follow suggestions from Sosa (2011) who claimed that testimony and inductive generalization through trial and error can provide a rationale for trusting any device to be reliable (not just epistemic devices), even though we do not understand how the device operates. I go further than his suggestions in the detail of my account and by making the suggestions directly relevant to dealing with the opacity of web technology.
I thank a reviewer for raising this issue.
Simpson goes on to argue that personalization in information filtering algorithms harms the objectivity of their results. Smart and Shadbolt (2016), however, provide some considerations in defense of personalization in information filtering.
Simon (2010) discusses in addition several features of Wikipedia that are transparent to the user’s key details about how pages are edited and by whom they are edited.
The extent of S’s understanding of the technology is about as much as one could glean from Google’s own explanation of how search works. Google’s explanation can be found online: https://www.google.com/intl/bn/insidesearch/features/voicesearch/ and https://www.google.com/intl/bn/insidesearch/features/search/knowledge.html. Accessed December 14, 2016.
I thank a reviewer for this example.
I thank a reviewer for raising this point.
https://www.google.com/intl/es419/insidesearch/tipstricks/. Accessed December 18, 2016.
The application of empirical methods to opaque devices is common and sometimes required in the use of complex computational instruments in science. Some computational processes are opaque to humans, because they operate so fast or with such complexity that human minds are unable to follow how the procedures produce subsequent states from initial states. Humphreys (2004) argued that this kind of opacity arising from computational speed and computational irreducibility does not block opaque computational models from being effective and useful in computational science. The models are useful because the models’ virtues can be achieved through trial and error despite opacity. For more detailed discussion on opacity in computationally assisted proofs and complex, agent-based modeling, see Humphreys (2004, 147–151) and Bedau (2014).
This method raises classical problems with justifying inferences based on inductive generalizations, but these problems are not my concern here. We form inductive generalizations naturally, and they are often useful. Furthermore, even if we cannot justify inferences from inductive generalizations, we can still corroborate or contradict particular outputs by comparing them against independent sources.
One example is ProPublica’s investigatory series on structural injustices perpetrated by bias in machine learning algorithms: https://www.propublica.org/series/machine-bias. Accessed, April 14, 2017.
References
Bedau, M. A. (2014). Testing bottom-up models of complex citation networks. Philosophy of Science, 81(5), 1131–1143.
Humphreys, P. (2004). Extending ourselves. Oxford: Oxford University Press.
Humphreys, P. (2009). Network epistemology. Episteme, 6(2), 221–229.
Lehrer, K. (1995). Knowledge and the trustworthiness of instruments. The Monist, 78(2), 156–170.
Miller, B., & Record, I. (2013). Justified belief in a digital age: on the epistemic implications of secret internet technologies. Episteme, 10(2), 117–134.
Pasquale, F. (2015). The black box society. Cambridge: Harvard University Press.
Simon, J. (2010). The entanglement of trust and knowledge on the Web. Ethics and Information Technology, 12(4), 343–355.
Simpson, T. W. (2012). Evaluating Google as an epistemic tool. Metaphilosophy, 43(4), 428–437.
Smart, P. R., & Shadbolt, N. R. (2016). The world wide web. In J. Chase & D. Coady (Eds.), Routledge handbook of applied epistemology. New York: Routledge Retrieved March 31, 2017, from https://eprints.soton.ac.uk/379968/.
Sosa, E. (2011). Knowledge: testimonial and instrumental. In Knowing Full Well, by Ernest Sosa (pp. 128–139). Princeton: Princeton University Press. https://10.1515/9781400836918
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Dahl, E.S. Appraising Black-Boxed Technology: the Positive Prospects. Philos. Technol. 31, 571–591 (2018). https://doi.org/10.1007/s13347-017-0275-1
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s13347-017-0275-1