“The principle of equal consideration of interests”: an instance of Mills’ fatal “idealized cognitive sphere" that captures real moral outcomes
The problem with “ideal” moral theory, says Charles Mills (Ideal Theory as Ideology, 2005), is that we can fail to capture a relevant moral reality that can be useful to all, especially those facing oppression or extreme advantage. This malignant sense of “ideal”, for Mills, concerns the use of “ideal-as-idealized models”, which rely on exemplary forms of actual moral conditions that deform essential features and distort essential constraints. One kind of this ideal-as-idealized model is the use of an “idealized cognitive sphere”, where the effects of otherness on social cognition are minimized or ignored. While Mills condemns this idealized cognition as a close accomplice to ignoring oppression in ideal moral theory altogether, the vast diversity and complex interactions of cognitive features with moral conditions cannot be denied; the effects of social differences may not be its most salient feature. Using Peter Singer’s “principle of equal consideration of interests”, crucial to the operation of his preference utilitarianism (Practical Ethics, 1979), I argue that Mills’ conception of an idealized cognitive sphere can be effectively employed in moral theory towards relevant outcomes. I shall examine, first, Mills’ view of an idealized cognitive sphere, identifying two features – ignorance of historical positioning and epistemic immodesty – that drive its fatality; second, how Singer’s principle and a derivative view employ a Millsian idealized cognitive sphere; third, how this principle, despite its use of this model in either variant, escapes these fatal features.
Mills’ use of “ideal” and the “idealized cognitive sphere”
The sense of “ideal” that I am concerned with is the most urgent sense defined by Mills in Ideal Theory as Ideology (2005), in which a moral theory employs exemplary moral conditions that not only deviate from actual moral conditions P in an essential way but also ignore this deviation as necessary and irrelevant. Mills contrasts this sense of “ideal” with, first, a generic “ideal-as-normative" sense, broadly applicable to any theorizing with P that appeals to normative value-systems, and second, an “ideal-as-descriptive" sense, which preserves the essential character of P and simplifies its lesser features to reach a moral scheme functionally congruent with P (166). Ideal-as-idealized models are not useful towards P in that its exemplar-P does not retain the salience or relations of its original features, making the ideal moral theory operate on distorted parameters.
An “idealized cognitive sphere” (or, as I will call, “idealized cognition”), then, is one possible strategy among other distorting strategies that drive ideal-as-idealized models. It posits a mental world equally employed by all, where epistemic differences are not only unaffected by social advantage, but are dismissed as personal cases of difficulty or bias that can be ignored. Idealized cognition often works in parallel with what Mills identifies as “idealized capacities” and “silence on oppression” (168); as I understand, cognition is the basis by which capacity can be assumed and does not operate solely on whether something can be achieved. By this distinction, the cognitive sphere concerns itself with the mechanisms by which moral decisions are made, not primarily their outcomes.
With this, I identify two features of idealized cognition that seem to prove it fatal to capturing actual moral conditions:
Ignorance of historical positioning. An idealized cognitive sphere fails in that it does not situate agents within a social order that has advantages or disadvantages meaningful to, and as a direct result of, historical inequality. The changing stakes associated with contexts subject to time cannot be ignored.
Epistemic immodesty. An idealized cognitive sphere is unaware of its own limitations, leading to false claims about ability, value systems, and scope of development. "Lack of epistemic modesty” was a point of criticism noted against SInger by Eva Feder Kittay (404; The Personal is Philosophical is Political, 2010); I believe a Millsian sort of idealized cognition involves this as well, to which Singer, in my view, escapes.
These “fatal features” will be used to dissect Singer’s grounding principle of equality as an exception to Mills’ critique of idealized cognition.
Singer’s principle and its use of an idealized cognitive sphere
We turn now to Peter Singer’s preference utilitarianism, grounded in the “principle of equal consideration of interests” (which I will call “PECI”, for short). To draw a tangible scope, I will focus on the kind of cognitive sphere implicated by PECI, not of Singer’s kind of utilitarianism in its entirety; Singer’s broader work may draw moral conclusions opposing my own, but I do not regard these as certain signs that my position has failed, nor will it be an obstacle to finding merit in PECI itself. I will, first, describe the principle, then, after establishing Singer’s vision of how it ought to be employed, offer a different use (the “participant view”), and last, attempt to show that PECI – in both views – uses Mills’ idealized cognitive model.
PECI, is, as follows: the pressures exerted by the like preferences of affected actors in a moral scene are to be considered equal, where, in assigning moral weight to possible outcomes, the measure of utility gained or lost is not tied to the prior position of any actor but to the value schemes put forth through these preferences. Singer’s example illustrates this in simple terms:
... If only X and Y would be affected by a possible act, and if X stands to lose more than Y stands to gain, it is better not to do the act. We cannot, if we accept the principle of equal consideration of interests, say that doing the act is better, despite the facts described, because we are more concerned about Y than we are about X. What the principle really amounts to is: an interest is an interest, whoever’s interest it may be. (20)
So, while the desire by a certain actor for their preference to be realized is divested from the moral scene, the preference itself remains as a player in determining the best course of action. Inspired by classical utility, Singer often uses the feeling of physical pain and its relief as a relevant example (21-22, 53): actors can be safely assigned a preference for no pain, lesser pain, or the maximal relief of existing pain, with the idea of pain being a moral offense in general, to be eased as much as possible, and with the agreeable property that any suffering or relief it involves cannot be transferred from one body to another. This is a curious thing: that, in Singer’s most intuitive case, an interest is likely self-serving or self-preserving, yet the moral scene it must participate in is a social one, where even minimal consideration by an actor for other actors is likely to emerge in their own interests, even if each weighed equally. I understand that this is where complications arise: Singer seems to assign the greatest deliberating power of a given moral problem to an unnaturally divested onlooker. He has no preference to contribute to the course of action himself, and is not only absent in the sense that he cannot volunteer a preference, but also cannot affect the moral weight of any of its parts. I will call this the “invisible jurist view”.
Here, I will ask that this view for PECI be held less dearly, at least for a moment; we will retain the core principle but employ it in a more transparent way, which I will call the “participant view”. The participant view argues that, while the equal consideration of all interests at play remains in effect, agents may exert interests that are beyond the self, using their moral stake to acknowledge and support others. Agents not only hold interests but exert them to contribute to the final course of action. Within this view, I will identify four kinds of preference:
The classical Singerian type of interest, where preference does not extend beyond self-preservation or self-gain; in a group where recourse must be unequally divided, and these are the only kind of interests at play, the equal weighing of these interests will result in benefit to the least advantaged or the greatest overall advantage to the group, if these interests persist as self-serving. Singer’s example of prioritizing urgent care to a doctor, for instance, such that they may assist others in need, is still reliant on the doctor’s interest being to secure their own health, weighed equally against the interests of each wounded actor to also secure their own health.
Empathetic preference. One’s own interest may acknowledge the interests of others, even those that cannot coexist with maximal self-gain; this may lead to X extending a voluntary benefit to Y, where X might otherwise choose to retain or gain this benefit for themselves. Using Singer’s example of the wounded doctor, the doctor feels urgency towards the suffering of others, so he may prefer to allocate more supplies to them than would be available if he had treated himself to the fullest extent. In this case, the doctor is choosing to defer his own maximal health – perhaps using makeshift bandages rather than the more complex dressings available – in favour of giving the other wounded a better treatment option.
Altruistic preference. X’s interest is to further Y’s interest, at X’s expense, but not to X’s demise; X remains the same as in their prior position, having given their potential gain to Y, but keeping any existing advantages. In this case, the wounds of the doctor and the others are equally non-fatal, with adequate supplies for all, yet he does not treat himself first, preferring to prioritize the others instead.
Self-sacrificial preference. X’s interest is to further Y’s interest to the maximal extent, in that X will defer all their existing or potential advantage to support Y. In this case, the doctor and the others equally require urgent care; supplies are not enough to save everyone. He prefers to forgo all treatment and supplies for himself, using his remaining capacity to treat the others. The doctor does not survive, but the others do.
These kinds of preference proceed from most-to-least self-central. The participant view aims to capture the social nature of preference when used to weigh outcomes that maximize utility among a group; if one imagines an interest (in PECI) as a single ballot placed towards pursuing a course of benefit, that ballot may only stand as one vote, but the content of the ballot allows for the voter to refer to candidates other than themselves. I will admit that my view also implies an option unlisted here, which is to prefer inaction or active disadvantage for everyone including oneself, but this is beyond the scope of this discussion.
My task remains to connect Mills’ idealized cognitive model to the cognition implicated by each view. We will recall that, for Mills, an idealized cognitive sphere entails the following:
It is shared by all actors. I will add that Mills uses the term “social transparency” (169); I take this to mean that social experiences are mutually understood as atomic cases generated by individuals acting in parallel. These experiences are also well-known among all actors.
It ignores the cognitive effects of oppression, historical and ongoing.
Difficulties it encounters can be ignored as outlying cases of bias or self-interest.
For PECI in general, this kind of cognitive model is quite useful. We do not have to contend with the messiness of an interest being unknown, nonsensical, or having historical roots to dissect; an interest is an interest is an interest, immediate and reasoned and personal. Transparency allows for interests to be common knowledge among those in a given moral scene, while the effects of oppression and personal inclination are neatly packaged together as one and the same. How PECI still employs this cognitive model despite the two variations of how it is presented – Singer’s view and my own – is similar:
For PECI as used in Singer’s invisible jurist view, the idealized cognitive sphere is non-negotiable; if interests are self-central, the possible benefit (and best outcome) at stake can be derived easily from their intent, which is always in favour of maintaining or gaining benefit to the actor that asserts it. Bias itself is no longer an epistemic problem, as all interests are biased, and predictably so; similarly, the effects of oppression are not a separate mechanism, but can be included in any interest. Transparency is also necessary: if interests are to be equally considered, they must be equally known to the jurist, or at least generally accessible to all who hold stake in a moral scene.
It is just as present in the participant view. Here, interests can be more socially positioned, but even this social dimension must rely on the interests of others being knowable, or at least, understood; oppression and its legacies are known as well, though this demand is less urgent, as interests being self-asserted – even in cases where the interest itself conveys no benefit to the asserter – involve, like in Singer’s view, natural bias and can express any cognitive effects from marginalization.
Yet, how PECI's use of a Millsian idealized cognitive sphere – which, we recall, is a part of ideal-as-idealized-models responsible for severely distorting the actual workings of moral scenarios – still results in rather undistorted theorizing is not yet clear. The following segment argues for the resilience of PECI in both the invisible jurist view and the participant view in the face of the two “fatal features” held by Mills’ ideal cognitive sphere.
The resilience of PECI to the fatal features of an idealized cognitive sphere
I will quickly revisit the two “fatal features” of a Millsian idealized cognition that I have identified: a lack of historical positioning, which drives the erasure of marginalization as a temporal force, and epistemic immodesty, where a lack of self-awareness by the cognitive model means that its limitations cannot be relied on to identify cognitive scope or diversity. These problems I deem “fatal” because they appear to be the forces most distorting when used in moral theorizing. How, then, does PECI use an idealized cognitive model that both satisfies Mills’ criteria and does not fall prey to these features?
I answer that, while PECI uses an idealized cognitive model, the main operative mechanism by which it participates in moral deliberation – preference – can wholly account for the social, personal, and historical forces affecting cognition. These effects will express themselves in the content of the preference and are not dependent on whether the agent themselves choose to do so. Forming a preference does not require one to know and respond to the effects of oppression, past or present, nor does it pose a threat to the preferences of others being in play if all are to be considered equally. We will refer again to the example of the wounded doctor: in the jurist view, he and the others prefer only their own pain be improved, with the jurist using the relief of suffering to identify the utility gained; she allots more care to the doctor so that he may treat the others after him. Note that the preferences involved, while all contending with one’s own suffering and its relief, sense its urgency in personal and diverse dimensions. Even if equally wounded and equally in pain, the others may have faced prior detainment or war, informing their preference with the desire for assured safety, while the doctor, who may have in fact been among the attacking forces responsible, informs his preference with medical expertise, knowing that he can, at least, make effective use of his tools. The participant view is similar, extending the possible pressures in preference-cognition to social ones – charity, repayment, reparations, and the deservedness or vulnerability of others to which we assign duty in ourselves.
To address these two fatal features more specifically: PECI appears to involve historical positioning, rather than ignore it entirely; preferences can express patterns caused by the effects of disadvantage, even if not personally realized. If instead of wounds, the moral scene involved giving a basket of gourmet cookies and fruit to either of two equally hungry groups, one may prefer the foods as a familiar family favourite, while the other may prefer them due to a historical lack of access to luxury foods, making them novel, even if current nutritional needs are generally met in both groups. In the participant view, Group A knows Group B is hungry as well, possibly preferring to give the basket to B regardless of whether they know the full extent to B’s experience of scarcity, as they, in their advantaged habits, do not see this basket as a rare chance to enjoy gourmet treats. Epistemic modesty is also preserved by PECI; individual epistemic positions and expressions of marginalization are embedded in the content of preferences. Since they are given equal weight in moral deliberation, contradicting preferences are resolved, in the jurist view, by the impartiality of the jurist, who will determine the course of maximal utility based on the point of conflict itself; in the participant view, actors may delegate some of this utility to the learning process whereby experiences of marginalization and their effects can be conveyed to dominant groups and their situatedness better understood through this process.
Conclusion
In this paper, I have aimed to show that Peter Singer’s grounding principle of equality, used in his preference utilitarianism (Practical Ethics, 1997) – called the “principle of equal consideration of interests” (PECI) - is resilient to the distorting forces of an “idealized cognitive sphere” described by Charles Mills (Ideal Theory as Ideology, 2005), despite the principle employing this very kind of cognition. First, I refer to Mills’ account of an idealized cognitive sphere, noting its place in ideal-as-idealized models and identifying two features that make it fatal to capturing a descriptive moral reality. I then turn to Singer’s account of PECI, exploring the implications of his view, which I modified towards a second, more socially-situated view. Both views, as employers of PECI, are shown to satisfy Mills’ criteria for an idealized cognition, but, I argue, they do not exhibit the distorting forces of either fatal feature. The result is that PECI, despite employing an idealized cognitive model, shows that it is a robust tool for more true-to-life moral theorizing, and, by extension, can confer descriptive promise to idealized others.
References
Kittay, Eva Feder. “The Personal Is Philosophical Is Political: A Philosopher and Mother of a Cognitively Disabled Person Sends Notes from the Battlefield.” Cognitive Disability and Its Challenge to Moral Philosophy, May 14, 2010, 393–413. https://doi.org/10.1002/9781444322781.ch22.
Mills, Charles W. “‘ideal theory’ as ideology.” Hypatia, vol. 20, no. 3, 2005, pp. 165–184. Summer 2005, https://doi.org/https://doi.org/10.1353/hyp.2005.0107 .
Singer, Peter. Practical Ethics. Cambridge University Press, 1979.