Knowles’ and Ditto’s “Warm” View of Human Reasoning:

Plato (left) and Aristotle (right)

Eric Knowles and Peter Ditto endorse what they call a “warm” view of human reasoning. This is a “hybrid view” that recognizes that human reasoning is “necessarily both” “’hot’ (rooted in motivations and emotions) [and] ‘cold’ (rooted in cognitive operations).” (15). In their estimation, this dichotomy between hot and cold reasoning — what they also refer to as the “preference-principle dichotomy” — “oversimplifies human psychology.” (4). In “Preference, Principle, and Political Casuistry,” Knowles and Ditto present empirical evidence for their favored warm, hybrid view and attempt to explain away this preference-principle dichotomy. 

Knowles and Ditto begin by noting that the attribution of principle is, inter alia, “a lay-psychological hypothesis concerning the causes of another’s behavior.” (3). But the behavior with which they are particularly concerned is moral and political judgment. Thus, they raise the question: “When it comes to basic judgments of right and wrong, do people (educated adults, at least) reason dispassionately from abstract moral principles (Kohlberg, 1969), or do they offer principles as rationalizing cover for their emotion-based intuitions (Haidt, 2001)?” (4). 

A hot explanation of these judgments offers a “preference-driven account” where the “causal factor underlying [a] political position” is something like an “emotional preference,” a “simple desire” or just “personal self-interest” (or “group-based interests”). (6-7). While a cold explanation “frames a political position as deriving [logically] from general intellectual or moral principles.” (7). [Although, there are also these people, who seem to be “cold types” but who see moral principles as “at best crutches a morally sensitive person would not require.”] Knowles and Ditto reject both explanations in favor of a view which sees moral and political judgment as caused by both underlying preference and general intellectual principles. They think that the preference-principle dichotomy assumed by both of these explanations originates in certain psychological biases — specifically, the actor-observer bias (i.e. “the tendency of actors to see their own behavior as responsive to situations, but for observers to see the same behavior as reflecting the actor’s dispositions”) and the “perceivers’ motivations… to hold a position view of themselves, as well as of others who share their attitudes and group membership.” (12). 

Having rejected the preference-principle dichotomy, Knowles and Ditto suggest a hybrid view in which “affective and motivational factors can influence judgments…by affecting the cognitive processes that underlie them [but also] cognitive processes constrain and shape affective influences on judgment.” (15). “People do not believe whatever they want [to] believe simply because they want to believe it.” (15). Instead, we are “clearly sensitive to the plausibility of our beliefs.” (15). And we work to maintain an ‘illusion of objectivity’, “essential” to producing “genuine belief.” (16). 

Since we care about possessing accurate and objective beliefs, the effects of our preferences on our beliefs must happen subconsciously and without awareness. Thus, Knowles and Ditto suggest that we engaged in an “implicit casuistry” in which, from case to case, “individuals unwittingly select principles that happen to provide intellectual justification for preferred conclusions.” (20). This happens because “affective preferences operating in a particular case guide reliance on general principles in such a way that selectiveness of the choice of principles is obscured from the reasoned.” (20). This process, in turn, shapes “a host of attitudes about controversial social and political issues.” (21). So, while “our political and moral choices may be experienced as principled,” they are, nevertheless, “shaped, if not determined,” by our “interests or preferences.” (4). 

For example, “one subtle – but effective – way that people skew judgments towards preferred conclusions is by ‘shifting the standards’.” (17). In one study (Norton et al., 2004), “male subjects faced with a decision about who to fire for a stereotypically male job (in the construction industry) were found to rank education as a more important hiring criterion than job experience when make decisions about applicants of unknown gender, but reversed this ranking if faced with a decision in which a male applicant had less education but more experience than a competing female candidate.” (18). In another experiment, participants were faced with “mock hiring decisions for a peer councilor.”  When “the applicants’ qualifications were strong or weak,” they “exhibited no preference for White over Black candidates.” But when “qualifications were middling, [anti-Black] bias emerged.” (43). The point for Knowles and Ditto is that “participants rejected the Black candidate only when doing so could be rationalized in terms of a race-neutral principle” and that those participants “almost certainly experienced their decisions as driven by [such] principles.” (44). 

Why not add a Daily Show video at this point exposing hypocrisy?  After all, they note the Daily Show as evidence and their point is about the selective application of principles.  …. ok, I will: 

Vodpod videos no longer available.

Knowles and Ditto also provide evidence that people switch between deontic and consequentialist justifications “selectively” as a result of underlying preferences. For example, they describe an experiment in which “political liberals and conservatives” were faced with one of two variations of a trolley problem.  The “key modification was that… half of the participants were faced with a decision about whether to push a man named ‘Tyrone Payton’ onto the tracks to save ‘100 members of the New York Philharmonic’, while the other half had to decided whether to push a man named ‘Chip Ellsworth III’ onto the tracks to save ‘100 members of the Harlem Jazz Orchestra.” (25). The result was that “Liberal college students were significantly more likely to endorse consequentialist principles when the trolley dilemma involved Chip than Tyrone, whereas conservative students showed no hint of that effect.” (26). 

So, while they “would embrace the Biercian point that principled explanations are often little more than a ‘front’ for the expression of self-based, group-based, or ideologically-based preferences.” (50). Nevertheless, they admit that “even preference-based reasoning frequently bears some of the constraining features of principled judgment.” (50). “[O]nce a principle is invoked, even in the service of justifying a specific affective preference, it has the potential of influencing other judgments that are subjectively relevant to the principle.” (50). 

A followup study involving the trolley problem “showed just this kind of ‘carry-over’ effect.” (51). Revealing that “[p]articipants seemed to perceive a strong constraint to remain consistent in their use of moral principles across the two scenarios, even when their initial choice of principle was evoked by motivational factors.” (51). In fact, this sort of effect leads Knowles and Ditto to conclude that “[a]t the extremes of generality, motivationally invoked principles might even be able to achieve a kind of functional autonomy, becoming detached from the motivations that produced them.” (52). Thus, they find it “entirely plausible” that ideologies could achieve “internal consistency, philosophical support, and substantial validity.” (52). In which case, they conclude that “it misses the mark to say that all judgments made on the basis of that commitment are ‘unprincipled’.” (53). 

They also suggest that these studies support the conclusion that “our intuitions about fundamental moral questions are at least (if not more) ‘valid’ than the principles available to justify them” perhaps because of some Humean “moral sense,” which “divine[s] truly ethical behavior.” (53). They speculate that these intuitions may “reflect adaptive insights accumulated over the course of human evolution,” what they later call our “implicit wisdom.” (54). Thus, “the kind of casuistic reasoning highlighted here is not necessarily illegitimate simple because of its post hoc nature,” I assume because it reflects this “wisdom.” 

Alex the Parrot, world-renowned bird reasoner

While I find much of what Knowles and Ditto suggest compelling, I worry about what they seem to assume, i.e. that there are some causal, psychological processes which confer legitimacy upon their judgment-results, and others which do not. I have had hard time understanding what this would mean. I can understand how we might come to see a certain causal process as producing legitimate judgments — perhaps even doing so always — but it seems that this would be because the results were antecedently legitimate, not because the process conferred legitimacy. 

It seems to me that part of what makes our brains so spectacular is that they are so capable of learning — and, specifically, learning on the basis of a perceived correct result. Perhaps this sort of learning accounts for the “carry-over effect” described above as well as a principles “functional autonomy.” Moreover, as Knowles and Ditto note (although they do not give it considerable attention), “reasoned moral analysis” sometimes overrides “intuitive moral reactions,” perhaps this sort of analysis is the relevant kind of “moral learning” — a decidedly consequentialist sort of analysis (see Joshua Greene‘s “The Secret Joke of Kant’s Soul.”) As a result, perhaps future “intuitions” can be more consequentialistly defensible as opposed to being the artifacts of our evolutionary history. But any such claim at this point is entirely speculative. 

Nevertheless, the evaluative has a tendency to swallow up whatever it touches; we can re-evaluate our “intuitions” and values. As Nietzsche shows, even morality – which presents itself as the morality, forbidding the possibility of alternatives – can be subjected to evaluative appraisal. That is, we can ask whether morality is itself good. Perhaps some of our intuitions reflect “implicit wisdom” but, even in those cases, critical re-evaluation can give us much-needed assurance that are judgments are sound and comfort that we conduct ourselves as we ought to. We should be learning what judgments are morally sound instead of learning how to defend the judgments we already make as morally sound.


8 Responses to Knowles’ and Ditto’s “Warm” View of Human Reasoning:

  1. Portia says:

    Your descriptions of Knowles and Ditto were great, Jeff! It was very clear and easy to read. I don’t really have anything to add besides a small type-o ” to swallows.” Great job!

  2. Kam says:

    Jeff –
    Do you think that all people experience “intuition” and “moral learning” in the same way? Moreover, does your image of Alex the Parrot indicate you think these processes exist, at least in some form, in other species?

    – Kam –

  3. jrpote says:

    Thanks for the comments!

    Kam, I have no idea what “intuition” is. Other people talk about having them, so I assume their is some sort of phenomena there… but I am also skeptical. As for Alex, I am not sure there is any evidence that he engaged in “moral” reasoning, but he did seem to possess at least something a lot like logical reasoning. I tend to be suspicious of claims that a particular human capacity is entirely novel.

  4. Arsen says:

    hey bud. Procedural comment: I like your cites (bcs I might find myself writing on the subject later on). But I wonder if a regular user would need them / find them distracting. Might be worth bringing it up in class as a “convention” question.

  5. tae0000 says:

    Hi, it certainly is a great summary. One comment! Yes, our brain can learn, but learning may not be so easy.

    Some researchers claim that there are critical periods for certain type of learning. A rather blatant example is language acquisition. According to these researchers, learning a language has to take place during a specific period of time.

    So someone who had an ear infection during the period in which they needed to develop phonological processing skills do not have a chance to develop their phonemic awareness ability. As a result, auditory inputs they receive do not mean anything, so they lose ability to develop their language skills.

    This may be a bit of stretch, but once someone’s “ideology” is firmly built in their system, learning a new thing or chaning their minds may not be as easy as we think as this may require them to re-wire their synapses.

  6. Arsen says:

    quick typo find: “while the other half had to decided whether to push . . . .” “DecideD” should probably be” decide.”

  7. Amit says:

    The insight that human reasoning is “warm” — rather than merely either hot or cold at any particular moment — has firm grounding in neuroscience. The emotional and cognitive regions of the brain do not work in isolation. Rather, neuronal connections flow between the emotional regions (e.g., amydgala) and the cognitive regions (e.g., prefrontal cortex).

    Moreover, the insight that human reasoning proceeds from preferences to principles — that reasoning is “warm-hot,” rather than lukewarm — is also grounded in cognitive science. Emotions are powerful.

    First, emotions tend to be cognitively impenetrable. The emotion of disgust is so powerful that people refuse to eat fudge in the shape of feces even if they fully comprehend that it is edible, and refuse to drink out of a perfectly clean toilet bowl. The neuroanatomical explanation is that domain-general mechanisms have less influence have over domain-specific mechanisms than vice versa. That is, the wiring from cortical areas (e.g., the prefrontal cortex) to the emotional areas (e.g., the amygdala) is weaker than the wiring from the emotional to the cortical. In fact, some emotional responses are executed with no inputs from the cortex.

    Second, emotions tend to be “fundamental.” Sentiments are fundamental features of the human mind; they are of such ancient evolutionary lineage that they must have been available for exploitation by natural selection. Cross-cultural studies reveal that certain basic emotions are pan-human, i.e., they are expressed in most or all human societies: these include anger, sadness, joy, fear, shame, pride, disgust, and guilt. The emergence of the neurobiological substrate of emotions, such as the amygdala, preceded the growth of cranial capacity and expansion of the neocortex. Studies in comparative cognition demonstrate that chimpanzees and bonobos, our closest animal relatives, have rich emotional capacities, and that our neurobiological hardware for fear resembles that of birds and reptiles. If our emotive programs are leftovers from the cognitive software of common evolutionary ancestors that existed even prior to the emergence our own species, then emotions are not of recent evolutionary lineage.

  8. Brian says:

    I think we did see the same thing in this article: that the authors had a tendency to label things preferences that could have actually been principles. They kept presenting what they thought of as morally equivalent situations, then explaining how people with similar views reacted differently to these equivalent situations. However, I found a lot of these problematic as the people who were supposedly being inconsistent could appeal to a principle to show that the two situations were actually morally different.

    For example, they talked about how conservatives were more inclined to accept attacks on the Iraqi military to save American lives but that they were less ok with attacks on the American military to save Iraqi lives. Therefore, they saw them as having inconsistent responses that could not be explained by principle. However, I feel like conservatives would vehemently debate that these situations are actually morally equivalent and would appeal to a “just war” principle that they would claim as applicable only to the American military attacks. So maybe there is more then pure preference at work here, and I think that the authors themselves even acknowledge this.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: