Although I have had little prior exposure to studies in the History of Philosophy, the class readings last week had a familiar ring to them.
Descartes’ theory of the intellect and the will was brought to the attention of cognitive scientists through a series of papers in the early 1990s by Daniel T. Gilbert and colleagues (Gilbert, Krull & Malone, 1990; Gilbert, 1991; Gilbert, Tafarodi & Malone, 1993). The focus in this literature is on the following claim from the Fourth Meditation (56), with square brackets indicating an addition to the French version:
Now all that the intellect does is to enable me to perceive [without affirming or denying anything] the ideas which are subjects for possible judgments.
For Gilbert (1991, 108) the “Cartesian model”, according to which some form of understanding occurs without reference to truth or falsity and prior to any evaluation of veracity, set an historical precedent:
Borrowing liberally from the Stoic philosophers, René Descartes was the first modern thinker to formalize this notion by partitioning the mind into relatively active (controlled) and passive (automatic) domains. Comprehension, he claimed, was passive: Ideas impressed themselves upon the mind as physical objects might upon soft wax … Although having ideas was effortless and automatic, accepting or rejecting those ideas was not. Descartes considered the assessment of an idea’s veracity to be the operation of the voluntas – the active, conscious, wilful force of the psyche.
In the Fourth Meditation Descartes aims to bring together the view that God gave us our faculties and his conviction that in God he “can find no cause of error or falsity” (54). In particular, he claims that God “did not give me the kind of faculty which would ever enable me to go wrong while using it correctly” (54). Given that humans do go wrong, Descartes proposes that such errors result solely from an incorrect use of one’s will in making affirmations or denials about matters that one does not clearly understand. In those cases the correct response is to exercise one’s freedom to refrain from making such judgments.
In contrast, Spinoza held that “Will and intellect are one and the same thing” and thus he was led to “deny that a man does not, in the act of perception, make any affirmation” (Ethics, 49th Proposition). According to the “Spinozan model”, as explicated by Gilbert, all ideas (including erroneous ones) are accepted as true in the initial process of understanding, but may be judged false through a process of assessment which typically quickly follows.
Within cognitive science it is commonly held that the process of assessment, understood in a broad sense, utilizes a “tagging” system in order to indicate truth-values. Under the Cartesian model ideas are initially untagged and may be subsequently tagged as either true or false. Under the Spinozan model untagged ideas are treated as true but may be subsequently tagged as false; thus, assessing an idea to be false is more demanding than assessing an idea to be true, in that it involves an additional step of actively unaccepting the idea. Although the Cartesian and Spinozan models both incorporate two stages (“understanding” and “assessment”), and under ideal conditions yield the same end results for assessments of truth and falsity, they diverge for cases in which the process at the second stage is either (i) resource depleted or (ii) interrupted. The Spinozan model predicts that false ideas should be taken for true ideas, but not vice versa, whereas the Cartesian model predicts no such asymmetry.
A number of studies have sought to provide an empirical test of the two models. For example, in Experiment 1 of Gilbert et al. (1990) subjects read a series of statements and were later asked to judge whether they were true or false. Subjects were also asked to quickly press a button when they heard a tone (the “distractor task”), which on a third of the trials was sounded 750ms after the relevant statement was presented. The interruption had no effect on the correct identification of true statements (55% vs. 58%) but did significantly reduce the number of correct identifications of false statements (55% vs. 35%). Importantly, Gilbert et al. (1993) provide evidence that when assessment is interrupted subjects will not merely tend to report false statements as being true, but will be prepared to base consequential behavior on those statements.
Within the limits of a brief blog post it is not possible to summarize or evaluate the scientific arguments for and against the Cartesian model, but I hope to have provided a flavor of how it has been operationalized. The research is especially interesting because although the picture of cognition that emerges from contemporary science differs in many ways from the picture that is found in the writings of Descartes, his view that we entertain “ideas” without taking them to be true (or false) has remained the common wisdom in science and philosophy. Although we may not be aware that we are treating the ideas as true, the issue is not that the stimulus is subliminal, that the ideas are non-conscious, or that our judgments are biased by uncontrollable heuristics and processing constraints.
The scientific literature raises two sets of questions in my mind. First, is the “Cartesian model” faithful to the philosophy of Descartes, and are Gilbert’s historical claims accurate? For example, precisely what do “passive” and “active” mean for Descartes, and could a Spinozan intellect reasonably be said to be affirming the ideas in a way that Descartes denies? Second, assuming for the sake of argument that the Spinozan model (which for our purposes we do not require to be faithful to Spinoza) is the correct one, does it fatally compromise Descartes’ position that God is not the cause of error or falsity? One worry is that the tagging mechanism under the Spinozan model of unaccepting an idea fails to allow for the possibility of refraining from taking an idea to be true or to be false, as is the case when judgment is suspended under the Cartesian model. However, it is not difficult to supplement the account by introducing, for example, an “uncertain” tag. Given that caveat, if ideas are affirmed by the intellect by default, and their status remains unchanged until the agent exercises their freedom to deem them not true, then should Descartes hold that we are any less responsible, and God any more responsible, for error or falsity?