Abstract
The articulation of an overarching account of scientific explanation has long been a central preoccupation for the philosophers of science. Although a while ago the literature was dominated by two approaches—a causal account and a unificationist account—today the consensus seems to be that the causal account (in one of its forms) has won. In this paper, I challenge this consensus and attempt to revive unificationism. More specifically, I aim to accomplish three goals. First, I add new criticisms (partly based on historical episodes) to the standard anti-unificationist arguments, in order to motivate the need for a revision of the doctrine. Second, and most importantly, I sketch such a revised version. Then I argue that, contrary to widespread belief (and in agreement with a small minority), the causal account and this revised unificationist account of explanation are compatible. Moreover, I also maintain that the unificationist account has priority, since a most satisfactory theory of explanation can be obtained by incorporating the causal account (properly spelled out), as a sub-component of the unificationist account. The driving force behind this reevaluation of the received view in the philosophy of explanation is a reconsideration of the role of scientific understanding.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Friedman (1974, 15). There have been attempts to endorse, refine and correct the view by several others, in particular by Kitcher (1981 and 1989); see also Jones (1995a, b, 2012), Weber (1999), Schurz and Lambert (1994), Schurz (1999), Bartelborth (2002). The thought that an explanation-understanding-unification link exists has been entertained in the past by several great philosophers such as Kant and Whewell. See Morrison (2000, ch. 2) for a review.
Not everybody believes that the only way to increase understanding is to provide explanations; for such a dissenting view, see van Fraassen (1985, 642). Although I think this is wrong, here I can’t argue against this position.
I will also refer to these as ‘phenomena’, although I don’t intend to confine the discussion to the realm of observable (of the type ‘metals expand when heated’); regularities taking place at the microscopic level (e.g., the photoelectric effect) are also meant to be covered by the account.
Woodward (2011 and 2003, ch. 8) contain inventories of some of the main criticisms; Morrison (2000) discusses the notion of understanding in connection to unification in more detail. Neither, however, concentrates on Friedman’s distinction on which I focus here, between local and global understanding.
It will be spelled out in Diagram f* in sect. 7 and 8 below. As I will make it clear later on, I will be making use of the distinction between various kinds of unification (including the ontological kind), which has been noted before by several authors, including Redhead (1984), Maudlin (1996), Morrison (2000) and Mäki (2001).
Following others, I shall take it to be the case that explanation and understanding are the two sides of the same coin. Kim (1994, 54), for instance, writes: “The idea of explaining something is inseparable from the idea of making it intelligible; to seek an explanation of something is to seek to understand it, to render it intelligible. These are simple conceptual points, and I take them to be untendentious and uncontroversial.” For a recent dissenting voice, see Gijsbers (2013).
Wolfgang Pauli also entertained the thought. See Heisenberg (1971, 29)
I should say at least two meanings. As a referee pointed out, there are recent discussions of understanding-that, −how, −with etc. I’ll leave them aside for reasons of space.
The distinction was introduced by Friedman (1974), mentioned briefly by Kitcher and Salmon in their early work on explanation, discussed in passing by Barnes (1992), and totally ignored ever since. None of the recent discussions of explanation and understanding in science even mentions it. See for example Woodward (2011), Gijsbers (2013), Grimm (2013, 2014), Hindricks (2013), Khalifa and Gadomski (2013), Khalifa (2013), Newman (2013), Strevens (2004, 2013).
The derivations are of course between descriptions of phenomena (i.e., sentences), not phenomena themselves. Although I’ll speak loosely throughout the paper (as customary in this literature), the relevance of paying attention to the descriptions of phenomena will become crucial in section 7.
Moreover, other types of derivation are possible, e.g., of statistical nature, or by taking limits (i.e., the ‘asymptotic’ type of explanation explored by Batterman 2002.)
I use a capital letter for it for further easier reference. Friedman (1974) takes it as a starting point.
Traditionally, the question asks about the ‘epistemic gain’ (Kim (1994, 54)) offered by an explanation.
However, this idea can’t be pushed aside so quickly—as Friedman (1974) does—because of the following complication. It might be the case that everyone in a community claims that they have ‘understood’ (and thus have no further why-questions to ask) not just because they are intellectually-psychologically comfortable with EXS, but for a deeper reason: they may lack the necessary conceptual framework within which to conduct this further inquiry.
I ignore the difficulties raised by the uniqueness of causes.
As suggested, I include here (ignoring the obvious differences) the group of philosophers supporting the idea that mechanisms are what matters when it comes to explanation. The works by Machamer, Darden and Craver are exemplary in this respect.
Recall that Hempel’s own suggestion was that understanding is to be identified with the subjective notion of nomic expectability (1965, 327).
And even turn things upside-down, by accounting for the ‘because’ of causation on the basis of the ‘because’ of explanation, as Kitcher urges (1989, 477).
I take it to be expressed in Woodward’s 2011 Stanford Encyclopedia of Philosophy article. Friedman’s position is mentioned, but not discussed.
A subsequent requirement is that d be a logical derivation; Friedman—and, as we’ll see, Kitcher too—follow Hempel and the logical positivists in this respect. I should note, however, that although the view I will eventually espouse here is unificationist, in the end I will not rush to follow Friedman and Kitcher in regard to the nature of d; as we’ll see (sect. 9), in my account I will ultimately leave the nature of d open.
For instance, we want to explain the trajectory of projectiles in the air, as well as why gases expand when heated.
Despite what Diagram f may suggest, scientific research doesn’t stop when a unifier UEXS is found; scientists will presumably treat it as an object of further investigation (a ‘mystery’) as well, and it will set the agenda for future scientific research. So, in a sense (as a referee has pointed out), unificationism is threatened by some form of infinite regress, or—just like causalism—by the incapacity to reduce the number of why-questions: if we take this number to be initially infinite, then eliminating a finite number of them will leave us with an infinite number still, and this can’t count as a reduction. Here I shall set this difficulty aside; the assumption that the number of why-questions is infinite seems to me not strictly speaking false, but rather exaggerated.
This is a point where we’ll return later, when discussing the conjunction problem.
This point aims to address Morrison’s (2000, 26–29) criticism, to the effect that Friedman’s account of understanding is not entirely objective, as it depends, after all, on what happens to be the explanatory ideal during a certain historical period.
This is part of Morrison’s point that unification comes apart from explanation (Morrison 2000, 4).
The point has been made by Barnes (1992, 8).
As Friedman emphasizes, science has as its ultimate target ‘understanding the world’ (1974, 19).
See De Regt (2006) who discusses Salmon's complementarity thesis.
As we’ll see in some detail in section 7, from here Friedman goes in the direction of worrying how we count the explanans. We’ll return to this later, as the issue is extremely important for the purposes of this paper.
The set of such triples [EXS, d, EXD] corresponds to what Kitcher calls the “explanatory store” (1981, 512)
Kitcher articulates this idea carefully, and some of the specifics are worth reflecting on, but here I’ll leave them aside, as they don’t really matter for what’s next (detailed expositions of them are available; see for instance Roland 2008)
One should not find this surprising. Just by reading carefully Kitcher’s two relevant works (1981 and 1989), one notices that each time he mentions the notion, he agrees with Friedman. In fact, he doesn’t seem to have much to say about understanding per se; for example, in his 1981 paper, he discusses understanding briefly at the outset on, p. 508–9, and then gets back to it only at the very end, on p. 529, where he agrees with Friedman.
Note 17 in Kitcher’s 1989 paper cites Friedman (1974); it runs as follows: ““…our total picture of nature is simplified via a reduction in the number of independent phenomena that we have to accept as ultimate” (Friedman 1974, 18). There is an interesting recapitulation here of T. H. Huxley’s summary of Darwin’s achievement. “In ultimate analysis everything is incomprehensible, and the whole object of science is simply to reduce the fundamental incomprehensibilities to the smallest possible number.” (Huxley 1896, 165)”
The italicized clause is important, since I’m not claiming that Kitcher’s theory is reducible to Friedman’s. For reasons of space I can’t discuss, as I warned above, Schurz and Lambert’s (1994) version of unificationism, which, unlike Kitcher’s, does take an account of understanding to be of paramount importance.
From (PV/T = constant), by taking Avogadro’s law into account, one derives the ideal gas law (PV/T = nR), and this quantity is constant for a given amount n of moles of gas, since R is the universal gas constant. As a result of the derivation of the ideal gas law within the kinetic theory, we have nR = NkB.
This aspect is by no means new. It was signaled in several careful and detail-oriented philosophical analyses of unification and reduction in science. In addition to the recent reference to Morrison’s book (2000, 2–3), earlier work by Sklar (1993) and Schaffner (1967) (as well as by others, including Feyerabend and Kuhn), should be mentioned. I’m not aware of any place where either Friedman or Kitcher discuss this ‘enhancing’ aspect in their work.
This is the more cautious version of the stronger claim I have in mind, that unification always leads to this enhancement—more precisely, as we’ll see, the right kind of unification, of the ontologically-reductive type. What follows are plausibility considerations in favor of the stronger claim, to be developed later on; the examples here are only illustrative.
D1 states that ‘S1 explains S2 iff S2 ∈ conκ (S1) and S1 reduces conκ(S1)’, and an interesting consequence of it, proved by Kitcher (1976, 209), is that only so-called ‘κ-atomic’ sentences can explain. I shall explicate some of this formalism in the benefit of completeness, but I stress that these details are not needed to understand the subsequent discussion here. A sentence S is ‘κ-atomic’ if it has no (κ-)partition, i.e., if there is no pair of sentences {A, B} such that A and B are acceptable independently of S, and the conjunction A & B is logically equivalent to S. κ is the set of accepted law-like sentences by the scientific community at a certain point in time, and is deductively closed. The set ‘conκ(S)’ is the set of independently acceptable consequences of S. A given sentence S ‘reduces’ a set Δ iff the following relation holds: κ-card (Δ U {S}) < κ-card (Δ). For a set of sentences M, the term ‘κ-card (M)’ denotes the κ-cardinality of M. Friedman stipulates (naturally) that κ-card (M) = inf card (Γ), where Γ is a κ-partition of M. As above, a κ-partition of a set of sentences M is a set of κ-atomic sentences which is logically equivalent to M (it is assumed that such a κ-partition exists for every set M).
There is of course more, e.g., absolute space and absolute time, but these are irrelevant here.
A consequence of all this is that Maxwell’s unificatory achievement mentioned above should not, on the view I propose here, count as genuinely explanatory. I agree with the referee who raised this issue that this sounds a bit counterintuitive, but I’m prepared to bite the bullet and point out that there is a sense in which the discovery of the nomological connections entangling electricity and magnetism only partially increased our understanding of the world—rather, it raised another question, as to why are they so entangled—to be answered later, by Einstein’s STR. However, the other, related, major Maxwellian idea, that light is (a form of) electromagnetic radiation, does count as explanatory, as it should, since it involves ontological reduction. The same referee has also raised the issue of other cases traditionally described as unifications (Mendelian genetics, Darwinian evolutionary theory, the theory of the chemical bond, etc.), and the verdict to be given is similar: the central question to ask when analyzing them (task which unfortunately can’t be carried out here) is whether ontological reduction is present. I suspect it is, in most of these cases; but if it is not, then, although they are still recognized as unifications, the view I support here will not deem them explanatory (despite the impression to the contrary).
As is immediate, the phenomenon in question here is in fact a member of a larger class of phenomena, each of them derivable from the same unifier.
And this is of course Friedman’s (1974, 19) worry: “How our total understanding of the world is increased by replacing one puzzling phenomenon with another”?
As Gijsbers (2007, 489–491) cogently argues, such problems continue to plague standard unificationism.
I should emphasize that Salmon’s ecumenism is based on the belief that the causal approach and unificationism are on a par, while I believe that they are not. Moreover, although I suppose that Strevens’ (2004) kairetic account can be called ecumenical, it explicitly prioritizes causation. Furthermore, what I’m saying here is also different from Kitcher’s previously mentioned ecumenistically-inclined idea that “the ‘because’ of causation is always derivative from the ‘because’ of explanation.” (1989, 477)
References
Ahmed, A. (2014). Evidence, decision and causality. Cambridge Univ. Press.
Barnes, E. (1992). Explanatory unification and scientific understanding. In D. Hull, M. Forbes, & K. Okruhlik (Eds.), PSA 1992 (Vol. 1, pp. 3–12). East Lansing: Philosophy of Science Association.
Bartelborth, T. (2002). Explanatory unification. Synthese, 130, 91–107.
Batterman, R. W. (2002) The devil in the details. Oxford Univ. Press
De Regt, H. W. (2004). Discussion note: Making Sense of Understanding. Philosophy of Science, 71: 98–109.
DeRegt, H. W. 2006. Wesley Salmon’s complementarity thesis: causalism and unificationism reconciled? International Studies in he Philosophy of Science 20, 129–147.
De Regt, H. W., & Dieks, D. (2005). A contextual approach to scientific understanding. Synthese, 144, 137–170.
Friedman, M. (1974). Explanation and scientific understanding. Journal of Philosophy, 71, 5–19.
Gijsbers, V. (2007). Why unification is neither necessary nor sufficient for explanation. Philosophy of Science, 74, 481–500.
Gijsbers, V. (2013). Understanding, explanation, and unification. Studies in History and Philosophy of Science, 44, 516–522.
Grimm, S. (2013). Understanding. In D. Pritchard & S. Berneker (Eds.), The Routledge companion to epistemology. Routledge.
Grimm, S. (2014). Understanding as knowledge of causes. In A. Fairweather (Ed.), Virtue scientia: Essays in philosophy of science and virtue epistemology. Special Issue of Synthese.
Heisenberg, W. (1971). Physics and beyond. Encounters and conversations (transation from German by A. J. Pomerans). New York: Harper and Row.
Hempel, C. (1965). Aspects of scientific explanation. New York: Free Press.
Hindricks, F. (2013). Explanation, understanding, and unrealistic models. Studies in History and Philosophy of Science, 44(3), 523–531.
Huxley, T. H. (1896). Darwiniana. New York: Appleton.
Jones, T. (1995a). How the unification theory of explanation escapes asymmetry problems. Erkenntnis, 43, 229–40.
Jones, T. (1995b). Reductionism and the unification theory of explanation. Philosophy of Science, 61, 21–30.
Jones, T. (2012). Unification. In S. Psillos & M. Curd (Eds.), The Routledge companion to the philosophy of science (2nd ed., pp. 97–106). New York: Routledge.
Khalifa, K. (2013). The role of explanation in understanding. British Journal for the Philosophy of Science, 64(1), 161–187.
Khalifa, K., & Gadomski, M. (2013). Understanding as explanatory knowledge: the case of bjorken scaling. Studies in History and Philosophy of Science, 44, 384–392.
Kim, J. (1994). Explanatory knowledge and metaphysical dependence. Philosophical Issues, Truth and Rationality, 5, 51–69.
Kitcher, P. (1976). Explanation, conjunction, and unification. Journal of Philosophy, 73, 207–12.
Kitcher, P. (1981). Explanatory unification. Philosophy of Science, 48, 507–31.
Kitcher, P. (1989). Explanatory unification and the causal structure of the world. In P. Kitcher & W. Salmon (Eds.), Scientific explanation (pp. 410–505). Minneapolis: University of Minnesota Press.
Kitcher, P., & Salmon, W. (Eds.). (1989). Scientific explanation, Minnesota studies in the philosophy of science, vol. 13, Minneapolis: Univ. of Minnesota Press.
Lange, M. (2002). An introduction to the philosophy of physics: Locality, fields, energy, and mass. Blackwell.
Mäki, U. (2001). Explanatory unification: double and doubtful. Philosophy of the Social Sciences, 31, 488–506.
Maudlin, T. (1996). On the unification of physics. Journal of Philosophy, 93, 129–44.
Morrison, M. (2000). Unifying scientific theories. Physical concepts and mathematical structures. Cambridge: Cambridge University Press.
Newman, M. (2013). Refining the inferential model of scientific Understanding. International Studies in the Philosophy of Science, 27, 173–197.
Quine, W. V. (1976). The ways of paradox and other essays (Revised and enlarged ed.) Cambridge: Harvard University Press.
Redhead, M. (1984). Unification in science. Review of The Unity of Nature by C. F. von Weizsäcker. British Journal for the Philosophy of Science, 35(3), 274–9.
Roland, J. (2008). Kitcher and the obsesive unifier. Philosophy and Phenomenological Research, 2, 493–506.
Russell, B. (1912). On the notion of cause. Proceedings of the Aristotelian Society, 13, 1–26. New Series.
Salmon, W. (1984). Scientific explanation and the causal structure of the world. Princeton: Princeton University Press.
Salmon, W. (1990). Scientific explanation: Causation and unification. Critica, 22(66): 3–21.
Salmon, W. (1998). Causality and explanation. Oxford: Oxford University Press.
Schaffner, K. (1967). Approaches to reduction. Philosophy of Science, 34(2), 137–147.
Schurz, G. (1999). Explanation as unification. Synthese, 120, 95–114.
Schurz, G., & Lambert, K. (1994). Outline of a theory of scientific understanding. Synthese 101, 65–120.
Sklar, L. (1993). Physics and chance. Oxford Univ. Press.
Strevens, M. (2004). The causal and unification approaches to explanation unified—causally. Noûs, 38(1), 154–76.
Strevens, M. (2013). No understanding without explanation. Studies in History and Philosophy of Science Part A.
Toulmin, S. (1961). Foresight and understanding. London: Hutchinson.
Trout, J. D. (2002). Scientific explanation and the sense of understanding. Philosophy of Science, 69, 212–33.
Van Fraassen, B. (1985). Salmon on explanation. Journal of Philosophy, 82, 639–651.
Weber, E. (1999). Unification: what is it, how do we reach and why do we want it? Synthese, 118, 479–99.
Woodward, J. (2003). Making things happen. A theory of causal explanation. New York: Oxford Univ. Press.
Woodward, J. (2011). Scientific explanation. In E. N. Zalta (Ed.), The Stanford encyclopedia of philosophy (Winter 2011 Edition). URL = http://plato.stanford.edu/archives/win2011/entries/scientific-explanation/.
Acknowledgments
I thank the two anonymous referees of this paper for really insightful suggestions and criticism. I am of course responsible for the final version of the paper.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Bangu, S. Scientific explanation and understanding: unificationism reconsidered. Euro Jnl Phil Sci 7, 103–126 (2017). https://doi.org/10.1007/s13194-016-0148-y
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s13194-016-0148-y