Abstract
In this paper, I argue that explanations just ARE those sorts of things that, under the right circumstances and in the right sort of way, bring about understanding. This raises the question of why such a seemingly simple account of explanation, if correct, would not have been identified and agreed upon decades ago. The answer is that only recently has it been made possible to analyze explanation in terms of understanding without the risk of collapsing both to merely phenomenological states. For the most part, theories of explanation were for 50 years held hostage to the historical accident that they far outstripped in sophistication corresponding accounts of understanding.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Hempel (1965) later placed an explicit restriction on content, demanding that D–N explanations invoke genuine laws. As of the 1948 paper, however, Hempel and Oppenheim remained optimistic that laws could be at least largely picked out by structural and syntactic markers.
Woodward’s presentation of the view was, admittedly, extremely approachable and non-technical. Issues pertaining to his view are explored in my ((Wilkenfeld 2013a, p. 64–67)).
This redirection is not wholly without historical precedent; as discussed in Sect. 5.2, the work of Peter Achinstein did start along similar lines.
Developed in greater detail in (Wilkenfeld 2013b).
Personal correspondence.
For a discussion of Hempel and Oppenheim on the link between explanation and understanding see Sect. 3.
It is not always clear just what the right sort of way is, and thus it is difficult to specify more precisely the exact causal connection these authors envision between explanation and understanding. (This is no fault of theirs, as the link to understanding was generally only of peripheral interest to their main projects.) The connection I argue exists will be articulated presently.
The weak requirement that an explanation only need to usable in an explaining-act if it were true is inserted to make sense of the following perplexing intuitions: we would say of a theory that we know to be false but which would account for some phenomenon that it is still, in some sense, a (bad) explanation. However, someone who presents that very theory to explain an occurrence of that phenomenon seems not to have explained it at all.
See later in this section for a discussion of the force of this modality, though a full accounting of the pragmatics of explanation-talk is beyond the scope of this essay.
Since the word “explains” appears within the scope of an intentional operator, its use does not make the definition at all circular. The content of B’s intention could be to cause understanding or simply to explain; in the latter case, what he would have would be a tokening of some conjugation or translation of “to explain” in the content of the intention itself. I do not know how precisely we specify the contents of intentions, but I take that to be a separate concern.
If one thinks ‘explains’ is not a success term, then ‘explains’ just means ‘engaged or attempted to engage in an explaining-act as defined by FE’; the remainder of this discussion could then be construed as providing an account of ‘successfully explains’. Similarly, if one thinks ‘explains’ does not necessarily denote intentional action, then the last clause of FE can be omitted in the definition of ’explains’; the remainder of the discussion could then be construed as an account of intentional explaining.
See my (Wilkenfeld 2013a, p. 49-50, especially n. 28) for a more extensive defense of this claim.
Similar remarks apply to the fact that only improved understanding is required—perhaps some readers will take what is really being defined to be ’partially explaining-act’, but on most accounts of understanding there is no obvious threshold that would do as a measure of success for an explaining-act.
Not too much need be said about the sorts of things that can interfere with an attempted explaining act’s failure. Attempts at explaining can fail because the audience has conflicting background beliefs, because the explanation is too computationally taxing, or because the audience is in the process of being shot in the head seventeen times. What matters for purposes of determining whether something is an explaining act was whether understanding was actually generated, and what matters for determining whether a bit of content is an explanation is whether it could have been used in an explaining act. The notion of “explanation” is admittedly thus somewhat permissive, where some explanations will only be usefully employable in very narrow contexts. However, as long as explanations are being evaluated relative to a particular context and set of background presuppositions, it will be generally clear whether a particular candidate explanation is the sort of explanation we would value. This strategy of beginning conceptually with successful explaining acts does face a challenge in specifying, for any particular context, under what circumstances something counts as a failed explaining act, but I take this to be a problem of secondary concern.
This is intended as a normative use of “normal”, not a statistical use. For an account of the sort of normality pictured, see Millikan (1993). The term “normal” is hers, though “natural” might be a better description of the envisioned characteristic.
It has been suggested to me that I cannot remain entirely neutral regarding this, if I want to maintain that my account is genuinely reductive. There is some worry that, at least in other areas, what ends up counting as the “right way” for a particular state (e.g. the perception of redness) to be caused ends up being individuated by its causing of that very state. However, while I would prefer not to give up on producing a reductive account of explanation, the production of a non-reductive but illuminating account would hardly be a disaster.
A weaker guarantor of veridicality in explanations comes from the fact that explanations are always representations. While this is not the place to develop a theory of representation, it is plausible that in order to count as a representation at all R must be veridical in some sense.
In the opening paragraph of her piece on the relationship between understanding and the facts, Elgin (2007) notes that “pretty plainly understanding somehow answers to facts. The question is how.” (p. 33).
This characterization is imperfect. One might think that understanding chess is relevantly analogous to understanding Newtonian mechanics, but is not best characterized as either understanding-why or understanding-how.
Relatedly, mathematicians often distinguish between explanatory and non-explanatory proofs (see Mancosu and Pincock 2012 for a discussion of this)—it is plausible that the distinction between the two is that the former produce the sort of cognitive achievement envisioned whereas the latter produce mere understanding-that. It is a potential strength of FE that it could shed light on this otherwise somewhat perplexing dichotomy, though arguing that it does so in precisely the right way is beyond the scope of this paper.
Note that the point is not that speakers would need a concept of e.g. argument structure in order to pick out D–N explanations, but that the proponent of such an account faces the added burden of providing a separate theory of the mechanism by which we pick out such recherché features.
See (Woodward 2003, Chapters 4.2 and 4.3) for a discussion of one strategy to relate D–N explanations to actual explaining acts.
It has been suggested that this reading is uncharitable, and that everything they say is compatible with the existence of a third use of ’understanding’ that is neither phenomenological nor rooted in explanation. I do not want to commit to any particular exegesis, but it seems fair to suppose that the failure of Hempel and Oppenheim and many of their intellectual descendants to discuss any other sort of understanding suggests that they at least did not place much significance on such an alternative.
Trout might only have meant to be discussing the phenomenal sense of understanding, but, if so, he does not seem to place much stock in the possibility of a more useful, objective sense.
Admittedly, their view of understanding a theory (as opposed to a phenomenon) does invoke a Hempelian model of explanation. It would be a relatively straightforward project to excise this; even as is, since the explanatoriness of the explanations does not play into their view, one could simply describe the relevant features of the Hempelian model without ever invoking explanation.
The term “tethered” is from Elgin (2007, p. 35), who attributes it to Plato. She argues against the even stronger requirement that understanding be factive.
For the remainder of the paper, I refer to the total view, consisting of UBEE and FE, merely as “FE”. The three theses—\(\hbox {UBEE}_{1}, \hbox {UBEE}_{2}\), and FE—can all be accepted or rejected independently of each other. However, when comparing my view to extant accounts of explanation, it is significantly easier to treat the theses as a package.
This assumes that I am correct about what diagnostic criterion philosophers actually use. See my (2013a, p. 60–71) for a defense of this claim by examination of examples.
Even Bas Van Fraassen’s account, while nominally pragmatic, falls into this category. See my (2013a, p. 60–69) for a more extensive discussion of counterexamples to Van Fraassen’s account and how they relate to FE.
It is possible this problem of strange beliefs leading to odd explanations could be patched by simply adding to Achinstein’s account a requirement that explainers have basically justified beliefs about what sorts of things cause understanding. I do not know whether such a move could be carried out in a manner consistent with Achinstein’s broader theory. The related problem of accounting for our ability to track others’ intentions would remain.
A commenter suggested that the real explanation of q in the sentence “the explanation of q is that p” is merely p, and thus the explanatory notion and any concerns about circularity drop out. That cannot be right though—on Achinstein’s account “the explanation of q is that p” is an explanation in exactly the same way “the cause of q is that p” is, yet it is clear that in the latter case the causal notion is an important part of the content of the explanation and not something that simply drops out. Salmon (2006, p. 149) similarly argues that Achinstein’s response is inadequate.
Moreover, I argue elsewhere (Wilkenfeld 2013b, p. 1013) that the dilemma that compels the authors to deny that models are representational is ultimately not compelling.
I deal with the interaction of FE, URM, and examples of explanatory asymmetry at greater length in (Wilkenfeld 2013a, Chapter 7).
I thank an anonymous reviewer for bringing this issue to my attention.
This assumes that not only do they understand incompleteness, but that they understand it in the very way my explanation purports to demonstrate it. A new proof or explanation of a familiar phenomenon can still generate new and improved understanding of it, and so would not pose a difficulty for the necessary connection articulated in FE. Notice that FE only demands that explaining acts generate improved understanding, and so a new explanation of an otherwise understood phenomenon would qualify.
I thank Stewart Shapiro for the suggestion that pragmatic halos could explain the intuitions in question.
References
Achinstein, P. (1983). The nature of explanation. New York: Oxford University Press.
Avigad, J. (2010). Understanding, formal verification, and the philosophy of mathematics. Journal of Indian Council of Philosophical Research, 27, 161–197.
Bechtel, W. (2008). Mental mechanisms: Philosophical perspectives on cognitive neuroscience. New York: Routledge/Laurence Erlbaum Associates.
Boumans, M. (2009). Understanding in economics: Gray-box models. In H. W. De Regt, S. Leonelli, & K. Eigner (Eds.), Scientific understanding: Philosophical perspectives (pp. 210–229). Pittsburgh, PA: University of Pittsburgh Press.
Bhushan, N., & Rosenfeld, S. (1995). Metaphorical models in chemistry. Journal of Chemical Education, 72(7), 578.
De Regt, H. W., & Dieks, D. (2005). A contextual approach to scientific understanding. Synthese, 144(1), 137–170.
Elgin, C. (2007). Understanding and the facts. Philosophical Studies, 132(1), 33–42.
Friedman, M. (1974). Explanation and scientific understanding. The Journal of Philosophy, 71(1), 5–19.
Hempel, C. G. (1965). Aspects of scientific explanation. In Aspects of scientific explanation, and other essays in the philosophy of science (pp. 1–504). Free Press.
Hempel, C. G., & Oppenheim, P. (1948). Studies in the logic of explanation. Philosophy of Science, 15(2), 135–175.
Kim, J. (1994). Explanatory knowledge and metaphysical dependence. Philosophical Issues, 5, 51–69.
Kitcher, P. (1989). Explanatory unification and the causal structure of the world. Scientific Explanation, 13, 410–505.
Knuuttila, T., & Merz, M. (2009). Understanding by modeling: An objectual approach. In H. W. De Regt, S. Leonelli, & K. Eigner (Eds.), Scientific understanding: Philosophical perspectives (pp. 146–168). Pittsburgh, PA: University of Pittsburgh Press.
Kvanvig, J. L. (2003). The value of knowledge and the pursuit of understanding. Cambridge: Cambridge University Press.
Lamont, M. (2009). How professors think: Inside the curious world of academic judgment. Cambridge, MA: Harvard University Press.
Lasersohn, P. (1999). Pragmatic halos. Language, 75, 522–551.
Leonelli, S. (2009). The impure nature of biological knowledge. In H. W. De Regt, S. Leonelli, & K. Eigner (Eds.), Scientific understanding: Philosophical perspectives (pp. 189–209). Pittsburgh, PA: University of Pittsburgh Press.
Mancosu, P. & Pincock, C. (2012). Mathematical explanation. Oxford Bibliographies Online. Philosophy. doi:10.1093/obo/9780195396577-0029. Retrieved Novenber 11, 2013.
Millikan, R. G. (1993). White queen psychology and other essays for alice. Cambridge: MIT Press.
Morgan, M. S., & Morrison, M. (1999). Models as mediators: Perspectives on natural and social science. Cambridge: Cambridge University Press.
Salmon, W. C. (2006). Four decades of scientific explanation (1st University of Pittsburgh Press pbk. ed ed.). [Pittsburgh]: University of Pittsburgh Press.
Scriven, M. (1962). Explanation, prediction, and laws. Minnesota Studies in the Philosophy of Science, 3, 170–230.
Trout, J. D. (2002). Scientific explanation and the sense of understanding. Philosophy of Science, 69(2), 212–233.
Van Fraassen, B. C. (1980). The scientific image. Oxford University Press.
Wilkenfeld, D. (2013a). Explaining and understanding. (Electronic Thesis or Dissertation). Retrieved from https://etd.ohiolink.edu. Retrieved March 31 2014.
Wilkenfeld, D. (2013b). Understanding as representation manipulability. Synthese, 190(6), 997–1016.
Wittgenstein, L., Anscombe, G. E. M., & Cumming, M. (1958). Philosophical investigations. Oxford: Blackwell.
Woodward, J. (2003). Making things happen: A theory of causal explanation. New York: Oxford University Press.
Ylikoski, P. (2009). The illusion of depth of understanding in science. In H. W. De Regt, S. Leonelli, & K. Eigner (Eds.), Scientific understanding: Philosophical perspectives (pp. 100–122). Pittsburgh, PA: University of Pittsburgh Press.
Zednik, C. (2011). The nature of dynamical explanation. Philosophy of Science, 78(2), 238–263.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Wilkenfeld, D.A. Functional explaining: a new approach to the philosophy of explanation. Synthese 191, 3367–3391 (2014). https://doi.org/10.1007/s11229-014-0452-z
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11229-014-0452-z