How the Tiger Bush Got Its Stripes: ‘How Possibly’ vs. ‘How Actually’Model Explanations

Abstract

1. Model-Based Explanation and The 'How Possibly'/ 'How Actually 'Distinction(s)The use of idealized models in generating scientific explanations is pervasive in the sciences; however philosophers of science have only recently begun to try to reconcile this practice with traditional philosophical accounts of explanation (e.g., Bokulich 2008; 2011).An important distinction for any model-based account of scientific explanation is that between 'how-possibly' model explanations and 'how- actually' model explanations. The notion of how-possibly explanation was first introduced by William Dray in the 1950s in the context of explanations in history. Dray conceived of how-possibly explanations as rival kind of explanation to deductive-nomological explanations, which he referred to as 'why-necessarily' explanations. He writes,Let us call these explaining why-necessarily and explaining how-possibly respectively. The two kinds . . . have different tasks to perform. They are answers to different kinds of questions. Explanations of the how-possibly pattem are often to be found in ordinary historical writing. The historian's problem is often to explain how some later event or condition could have come to pass in spite of known earlier conditions which give rise to contrary expectation. (Dray 1957, 161-62)There are three key characteristics to Dray's how-possibly explanations: first, they are distinctive kind of explanation; second, they are complete in themselves; third, their aim is to show that the explanandum event not have caused surprise (Dray 1957, 157).Dray illustrates this kind of explanation using the following example of the puzzlement caused by radio announcer's description of baseball game: It's long fly ball to center field, and it's going to hit high up on the fence. The centre fielder's back he's under it, he's caught it, and the batter is (Dray 1957, 158). As Dray notes, this description caused considerable puzzlement to the listeners who knew the fence was twenty feet high. The explanation, which spectators could have provided, is that there was platform with ladder for the scorekeeper, and the centerfielder scrambled up the ladder to catch the ball twenty feet off the ground. Dray concludes that the demand for explanation here was simply to show how it was possible and the explanation given can be considered complete. As some critics have pointed out, however, the explanation given in Dray's example does more than simply show the catch was possible; it gives the central mechanism by which the catch was actually made (e.g., Reiner 1993, 65).Subsequent discussions of how-possibly explanations typically treat them instead as explanations. Wesley Salmon, for example writes, a how possibly question does not require an actual explanation; any explanation not ruled out by known facts is suitable answer (Salmon 1989, 137). On this view, how-possibly explanation doesn't need to pick out the actual (incomplete) mechanism by which the event occurred, but it does need to be consistent with known facts. It is candidate for how-actually explanation. This is the sense of how-possibly explanation as it was revived in the 1990s by Robert Brandon in the context of the philosophy of biology.Brandon also defines how possibly explanations as potential explanations, none of whose explanatory premises contradict or conflict with 'known facts' (Brandon 1990, 179).' He is concerned more specifically with explanations of adaptations in evolutionary biology, such as Darwin's speculative explanation of how complex organ like the eye could have evolved by natural selection. Brandon argues that such an explanation of how trait could have evolved has cognitive value that is independent of whether or not it was the way the trait actually evolved. Unlike Dray, how-possibly explanations for Brandon are not rival to deductive-nomological explanations nor are they answers to fundamentally different kinds of questions. …