- Hume's Science of Human Nature: Scientific Realism, Reason, and Substantial Explanation by David Landy
In his bold and excellent book on Hume's scientific methodology, David Landy positions himself between the "Deductive-Nomological" reading, which explains particular phenomena in terms of empirical regularities, and the "New Hume" position, which considers empirical regularities to be the explananda and unknowable essences the explanans. Landy sides with the New Humeans, except that for him the essences, or "theoretical posits," are knowable. These essences become knowable, despite their being in principle unobservable, through the tool of "perceptible models." Landy argues that "theoretical posits are only possible via the deployment of a perceptible model," and he explains that perceptual models allow us to specify "determinate ways that the represented theoretical entity both resembles and differs from some perceptible object" (53).
The problem, however, is to specify the determinate way in which an object differs from a perceptible model when the difference concerns the imperceptible. Consider the cases that Landy identifies as the "two most important distinctions that Hume draws in the Treatise, the impression-idea distinction and the simple-complex distinction" (53). As for the latter, Landy maintains that it is not possible to perceive simples as such because experience is always and necessarily complex, and that, nonetheless, we can represent simples by taking complex perceptions as models. But how do we represent the simplicity of simple perceptions? Landy writes that "relative ideas as the New Humeans understand them are out" (53), precisely because they lack pictorial content. However, his own solution is not convincing either: what pictorial content can a perceptible model supply to the representation of simple perceptions? In places, he seems to say that simple perceptions are represented via their explanatory power. But then, New Humean essences too are represented, and thus knowable, in this way.
Moreover, Landy's defense of his claim that simple perceptions are unobservable is problematic. Having argued that the revival set of "simple idea" is a set of complex ideas, because that is all we can ever perceive, Landy briefly discusses the possibility that we might "focus" or "attend" to the simple elements of complex ideas and thereby perceive simple objects. Landy objects that the notions of focus and attention "are nowhere to be found" in Hume's theory of general representation (36). First of all, this is not true. Hume's discussion of the distinction of reason between figure and color is full of this language. Hume writes that we "view them in different aspects," that we can "consider only the figure," that we "tacitly carry our eye," that we "view its resemblance," that we can "still keep in our eye the resemblance" (T 18.104.22.168). Second, Landy himself relies on this language in his own interpretation of Hume's impression-idea distinction. He argues that, in establishing this distinction, Hume is "drawing his reader's attention to the phenomenal qualities by which each [perception] can be recognized: their force and vivacity" (44–45). Indeed, it seems that we can focus on the force and vivacity of our perceptions while ignoring their content and other relations, even though we never experience force and vivacity alone either. [End Page 350]
Finally, the "grounding" role Landy ascribes to his theoretical posits is questionable. Landy argues that the essences we discover ground phenomenal differences. He uses the following model: we pre-theoretically distinguish gold and lead, and then we discover that differences in atomic weights ground the phenomenal differences (47). The same, he claims, is true of the distinction between impressions and ideas: we distinguish them via phenomenal differences, and then discover that ideas are copies while impressions are mental originals. This difference is supposed to ground the phenomenal differences. However, whereas we can explain the observable differences between lead and gold in terms of their respective atomic weights, the relation between copy-original on the one hand, and strength of force and vivacity on the other hand, seems unintelligible. What, in the fact of being an original, explains...