Cimpian & Salomon (C&S) describe a process by which inchoate intuitions outputted by the inherence heuristic are developed by other cognitive processes into beliefs about psychological essentialism. There is much of interest in their proposal, but here I discuss the potential to move beyond this specific instantiation to a general model that may have wider application. To begin, I identify what I take to be the central insight of the authors, combine that with other insights drawn from cognitive science, and describe the features of the general model. To illustrate the general model, I apply it to belief in objective value. And I conclude by highlighting an implication of the general model that is relevant to comments made by the authors about moral reasoning.
The central insight of C&S is that psychological essentialism “emerges as an elaboration of the earlier, and more inchoate, intuitions supplied by the inherence heuristic” (sect. 4, para. 1). In claiming this, I assume the authors accept: (1) the existence of two systems of cognition as described by Stanovich and West (Reference Stanovich and West2000), (2) that System 1 (intuition) outputs inchoate (I take that to mean “undeveloped”) intuitions, and (3) that System 2 (reasoning) elaborates upon those to produce developed beliefs associated with psychological essentialism. But I suggest that the process they describe is only one instantiation of a more general process of interaction between undeveloped intuitions generated by System 1 heuristics and developed beliefs endorsed by System 2.
To fill out this general model of interaction between System 1 and System 2 further, let me identify some other insights. Kahneman (Reference Kahneman2002) describes intuition (and here let us assume the inherence heuristic is one source of such intuition) as “occupy[ing] a position – perhaps corresponding to evolutionary history – between the automatic operations of perception and the deliberate operations of reasoning” (p. 450). This comment prompts a question: Is intuition (e.g., the outputs of the inherence heuristic) closer to perception or reason? From what they say, I take C&S to assume that the functioning of the heuristic is closer to what we call reasoning. But what if its functioning is much closer to what we call perception?
C&S distance themselves from the possibility that the inherence heuristic could be an “innate module,” preferring to characterize the associated process as a “highly constrained developmental process” that generates “essentialist beliefs out of simpler cognitive parts” (sect. 4.2, para. 1). But what if these “simpler cognitive parts” are the outputs of a module? There are features of modules that fit well into the general model that I am advancing. For example, Fodor (Reference Fodor1983) describes the outputs of peripheral modules as mandatory and contends that the manipulations made inside these modules are relatively inaccessible to introspection (p. 55). The Müller–Lyer illusion is a good example of the operation of such a module (or set of modules). No detailed neurophysiological explanation of the illusion has achieved broad acceptance (Bertulis & Bulatov Reference Bertulis and Bulatov2001, p. 5), but presumably the retina receives information that corresponds to two parallel lines of equal length, and somewhere between the retina and the conscious experience of the illusion the information representing the lines is manipulated to represent lines of unequal length.
I invite C&S to reflect on the insights of Kahneman and Fodor when considering their proposal. Perhaps the functioning of the inherence heuristic is closer to perception than to reason, and its functioning is more inaccessible to introspection than they have assumed. If so, then they may agree that the inherence heuristic is one instantiation of a more general model of interaction between System 1 and System 2.
In this general model, undeveloped intuitions about represented objects, events, or states of affairs are produced by one (or more) System 1 heuristic. These undeveloped intuitions are then available to System 2, and System 2 generates developed beliefs about these objects, events, or states of affairs. Elsewhere, I have described this process as System 1 providing System 2 “with the concepts with which it thinks” and illustrated it by claiming that System 1 provides System 2 with pieces of a jigsaw, and System 2 then does what it can to construct a coherent picture with those jigsaw pieces (Wood Reference Wood and Nagasawa2012, p. 77). I suggest the inherence heuristic can be understood as an instantiation of this general model.
To give another instantiation of the general model (I call it the “jigsaw-piece” model) consider moral judgement. Mackie (Reference Mackie1977) correctly observes that “ordinary moral judgements include a claim to objectivity, an assumption that there are objective values” (p. 35). Furthermore, Mackie claims that “if there were objective principles of right and wrong, any wrong (possible) course of action would have not-to-be-doneness somehow built into it” (p. 40). Now, the question of the existence of objective value is beyond the scope of this discussion, but what is of interest here is Mackie's identification of “not-to-be-doneness” because the nature and origin of this “objective, intrinsic, prescriptivity” (p. 35) is considered mysterious by many philosophers. But the jigsaw-piece model removes the mystery and offers a way to understand the relationship between “not-to-be-doneness” and belief in objective value. An undeveloped “not-to-be-done” intuition is outputted from a System 1 heuristic (like the way the inherence heuristic outputs a inchoate intuition), and System 2 constructs a developed belief in objective value (again, like the way System 2 constructs a developed belief in psychological essentialism). And, the “not-to-be-doneness” intuition could be innate or set by social referencing for particular activities (just like the way the inherence heuristic works).
In conclusion, C&S claim that understanding the inherence heuristic will help avoid “is–ought errors” in moral reasoning (sect. 3.4). But if the general model described here is correct, and if it includes, as an instantiation of it, the process that produces belief in objective value, then it has implications for moral reasoning that challenges C&S's claim. Indeed, the general model challenges moral reasoning itself because the “not-to-be-doneness” intuition – that is the foundation of our belief in objective value and thus is behind the central assumption underlying ordinary moral judgement – may be an illusion, like the apparent objectivity of the different lengths of the lines in the Müller–Lyer illusion.
You have
Access
Cimpian & Salomon (C&S) describe a process by which inchoate intuitions outputted by the inherence heuristic are developed by other cognitive processes into beliefs about psychological essentialism. There is much of interest in their proposal, but here I discuss the potential to move beyond this specific instantiation to a general model that may have wider application. To begin, I identify what I take to be the central insight of the authors, combine that with other insights drawn from cognitive science, and describe the features of the general model. To illustrate the general model, I apply it to belief in objective value. And I conclude by highlighting an implication of the general model that is relevant to comments made by the authors about moral reasoning.
The central insight of C&S is that psychological essentialism “emerges as an elaboration of the earlier, and more inchoate, intuitions supplied by the inherence heuristic” (sect. 4, para. 1). In claiming this, I assume the authors accept: (1) the existence of two systems of cognition as described by Stanovich and West (Reference Stanovich and West2000), (2) that System 1 (intuition) outputs inchoate (I take that to mean “undeveloped”) intuitions, and (3) that System 2 (reasoning) elaborates upon those to produce developed beliefs associated with psychological essentialism. But I suggest that the process they describe is only one instantiation of a more general process of interaction between undeveloped intuitions generated by System 1 heuristics and developed beliefs endorsed by System 2.
To fill out this general model of interaction between System 1 and System 2 further, let me identify some other insights. Kahneman (Reference Kahneman2002) describes intuition (and here let us assume the inherence heuristic is one source of such intuition) as “occupy[ing] a position – perhaps corresponding to evolutionary history – between the automatic operations of perception and the deliberate operations of reasoning” (p. 450). This comment prompts a question: Is intuition (e.g., the outputs of the inherence heuristic) closer to perception or reason? From what they say, I take C&S to assume that the functioning of the heuristic is closer to what we call reasoning. But what if its functioning is much closer to what we call perception?
C&S distance themselves from the possibility that the inherence heuristic could be an “innate module,” preferring to characterize the associated process as a “highly constrained developmental process” that generates “essentialist beliefs out of simpler cognitive parts” (sect. 4.2, para. 1). But what if these “simpler cognitive parts” are the outputs of a module? There are features of modules that fit well into the general model that I am advancing. For example, Fodor (Reference Fodor1983) describes the outputs of peripheral modules as mandatory and contends that the manipulations made inside these modules are relatively inaccessible to introspection (p. 55). The Müller–Lyer illusion is a good example of the operation of such a module (or set of modules). No detailed neurophysiological explanation of the illusion has achieved broad acceptance (Bertulis & Bulatov Reference Bertulis and Bulatov2001, p. 5), but presumably the retina receives information that corresponds to two parallel lines of equal length, and somewhere between the retina and the conscious experience of the illusion the information representing the lines is manipulated to represent lines of unequal length.
I invite C&S to reflect on the insights of Kahneman and Fodor when considering their proposal. Perhaps the functioning of the inherence heuristic is closer to perception than to reason, and its functioning is more inaccessible to introspection than they have assumed. If so, then they may agree that the inherence heuristic is one instantiation of a more general model of interaction between System 1 and System 2.
In this general model, undeveloped intuitions about represented objects, events, or states of affairs are produced by one (or more) System 1 heuristic. These undeveloped intuitions are then available to System 2, and System 2 generates developed beliefs about these objects, events, or states of affairs. Elsewhere, I have described this process as System 1 providing System 2 “with the concepts with which it thinks” and illustrated it by claiming that System 1 provides System 2 with pieces of a jigsaw, and System 2 then does what it can to construct a coherent picture with those jigsaw pieces (Wood Reference Wood and Nagasawa2012, p. 77). I suggest the inherence heuristic can be understood as an instantiation of this general model.
To give another instantiation of the general model (I call it the “jigsaw-piece” model) consider moral judgement. Mackie (Reference Mackie1977) correctly observes that “ordinary moral judgements include a claim to objectivity, an assumption that there are objective values” (p. 35). Furthermore, Mackie claims that “if there were objective principles of right and wrong, any wrong (possible) course of action would have not-to-be-doneness somehow built into it” (p. 40). Now, the question of the existence of objective value is beyond the scope of this discussion, but what is of interest here is Mackie's identification of “not-to-be-doneness” because the nature and origin of this “objective, intrinsic, prescriptivity” (p. 35) is considered mysterious by many philosophers. But the jigsaw-piece model removes the mystery and offers a way to understand the relationship between “not-to-be-doneness” and belief in objective value. An undeveloped “not-to-be-done” intuition is outputted from a System 1 heuristic (like the way the inherence heuristic outputs a inchoate intuition), and System 2 constructs a developed belief in objective value (again, like the way System 2 constructs a developed belief in psychological essentialism). And, the “not-to-be-doneness” intuition could be innate or set by social referencing for particular activities (just like the way the inherence heuristic works).
In conclusion, C&S claim that understanding the inherence heuristic will help avoid “is–ought errors” in moral reasoning (sect. 3.4). But if the general model described here is correct, and if it includes, as an instantiation of it, the process that produces belief in objective value, then it has implications for moral reasoning that challenges C&S's claim. Indeed, the general model challenges moral reasoning itself because the “not-to-be-doneness” intuition – that is the foundation of our belief in objective value and thus is behind the central assumption underlying ordinary moral judgement – may be an illusion, like the apparent objectivity of the different lengths of the lines in the Müller–Lyer illusion.