Rational Explanation and the Role of Agency

Hempel’s model also struggles to accommodate explanations that hinge on human agency and intentionality. While he dismisses the idea that historical understanding requires empathy—the ability to see events from the perspective of historical actors—he does acknowledge that rational explanations, which account for the reasons behind actions, are important (Hempel, 1942, p.249). However, he insists that such explanations must still be reducible to general laws.

Critics argue that rational explanation constitutes a distinct form of historical reasoning. Dray (1957, p. 95) contends that human actions can often be understood through practical reasoning rather than strict causal laws. If a historian explains why Martin Luther initiated the Protestant Reformation, they might cite his theological convictions, the corruption of the Catholic Church, and the availability of the printing press. While these factors contribute to an understanding of Luther’s actions, they do not function as general laws governing all religious reform movements.

Quine’s critique of meaning and reference in Word and Object (1960) also raises problems for the covering-law model. Quine argues that strict logical criteria do not fix meaning but are instead shaped by indeterminacy and contextual interpretation (Quine, 1960, p. 26). If historical explanations rely on rational actors whose decisions are influenced by shifting linguistic and cultural contexts, then explanations become useless. This suggests that historical explanation, much like language, is fundamentally underdetermined rather than governed by absolute causal laws.

Causal vs. Narrative Explanation

Another fundamental problem with the covering-law model is its difficulty in accounting for narrative explanations. Historians often explain events by tracing their development over time, showing how one event led to another through a series of interconnected decisions and circumstances. This form of explanation does not necessarily rely on general laws but rather on a coherent reconstruction of historical sequences (White, 1973, p. 7).

For example, the outbreak of World War I is commonly explained by narrating the escalation. Tensions between European powers, the assassination of Archduke Franz Ferdinand, and the web of alliances that turned a regional conflict into a global war. While causal factors are certainly present, the explanation primarily functions as a historical narrative rather than a deductive argument (Carr, 1961, p. 45).

​Paul A. Roth’s How Narratives Explain (1989) presents a critique of Hempel’s model by arguing that narratives explain in their own right, not merely as illustrations of formal logical models.

Roth defines a narrative as a telling of something that happened, typically an account of the linkage of events leading to a certain outcome. Unlike scientific explanations, which invoke causal laws, narratives describe discrete events (Roth, p. 1). Roth describes history as nonscientific, being both idiographic and nomothetic—meaning it pertains to singular, non-recurring events. Roth critiques traditional approaches, such as those of Carl Hempel and Ernest Nagel, which attempt to fit narratives into deductive frameworks, and instead suggests that explanation arises from the structure of a story (Roth, 1989, p. 453).

Roth examines Arthur Danto’s claim that narratives require generalizations (464–467) and Hayden White’s view that historical narratives follow literary structures (460–463), but he finds both lacking. Roth concludes that narratives serve as paradigms for explanation, much like Thomas Kuhn’s scientific paradigms. Rather than just listing facts, narratives shape understanding by constructing meaning (Roth, p. 477).

The Case for a Pluralistic Approach

Given these critiques, a more pluralistic approach to historical explanation is necessary—one that acknowledges the value of general laws where applicable but also accommodates rational, narrative, and conceptual explanations. Dray’s alternative model suggests that history does not need to adhere strictly to the methods of the natural sciences but should instead develop its own explanatory frameworks suited to its subject matter (Dray, 1957, p. 110).

Furthermore, the role of historical interpretation must be emphasized. Unlike the sciences, where explanations are judged primarily by their predictive power, historical explanations are often evaluated based on their coherence, plausibility, and capacity to illuminate the past (Windelband, 1894, p. 13).

Conclusion

Hempel’s covering-law model set an ambitious standard for historical explanation, seeking to align history with the methodologies of the natural sciences. However, its rigidity has been widely criticized for failing to account for the uniqueness of historical events, the role of human agency, and the explanatory power of narrative structures. Quine’s indeterminacy thesis further undermines the plausibility of the model. A more adequate framework must incorporate rational explanations, narrative structures, and interpretive methods alongside causal analysis. The ongoing debate between positivist and interpretivist approaches to history underscores the need for a pluralistic methodology—one that respects the distinctiveness of historical inquiry while maintaining rigorous standards of explanation.