Ignorance and Decision-Making, Part Three
The Unenviable Epistemic Position of Surrogate Decision-Makers
This is the third part of a multi-part series on the significance of ignorance for decision-making. In the first part of the series, I discussed the notion that decision-making is constrained by the decision-maker’s relevant ignorance. In the second post, I addressed the various philosophical arguments that I have developed, both alone and in collaboration with my co-author, Professor Parker Crutchfield of Western Michigan University, in defense of the thesis.
I had initially been planning to follow-up last week’s newsletter post with a discussion of the experimental studies that Professor Crutchfield and I have run, and the empirical evidence we have discovered in support of the logical priority of the epistemic. However, it dawned on me that readers might be better served, might come to better understand the thesis and, especially, its connection to the problem of policymaker ignorance, if I first discussed the problems that relevant ignorance poses for surrogate decision-makers. I will return later this week or next with the promised discussion of our experimental studies and empirical evidence.
According to prevailing ethical standards, surrogate decision-makers, those individuals charged with making decisions – whether medical, legal, political, or other kinds of decisions – on behalf of other persons, unable or unwilling to decide for themselves, ought to base their decisions on their judgment of what the surrogated person would want, if they could express a judgment under the circumstances. Surrogate decision-makers should substitute their judgment for the surrogated person’s would-be judgment. [1] However, it sometimes, perhaps often, happens that this standard can be met only if a surrogate decision-maker first surmounts various epistemic burdens.
First, a surrogate decision-maker might be ignorant of some combination of the surrogated person’s pertinent values, beliefs, attitudes, or dispositions, knowledge of which is necessary if the surrogate decision-maker is to decide for the surrogated person as this person would decide. Most of the time, when we make personal decisions for ourselves, this knowledge is automatically available to us. We typically don’t need to reflect at length or in depth on the sort of person that we are before we can make a decision for ourselves. If she is to satisfy the prevailing ethical criterion, however, a surrogate decision-maker must somehow acquire adequately intimate knowledge of the surrogated person’s values, beliefs, etc. Of course, this knowledge is often difficult to acquire. It is not easy to peer into another person’s soul, especially that of an incapacitated or otherwise incommunicative person, who might even be a stranger to the surrogate decision-maker, to determine how they would decide if they were capacitated and communicative.[2]
Second, a surrogate decision-maker might be knowledgeable of a surrogated person’s relevant values, beliefs, attitudes, and dispositions, but be unable to translate this knowledge into a decision that the surrogated person would make if they could decide for themselves. The surrogate decision-maker may be ignorant of what the surrogated person’s patient’s values, beliefs, etc., imply about the decision this person would make for themselves under prevailing circumstances. Again, when we make personal decisions for ourselves, we do so in the context of our own lives and in (more or less) full, if often only implicit, knowledge of how we want our lives to proceed. However, for a surrogate decision-maker, this deliberation must be explicit in order to sympathetically substitute their judgment for the surrogated person’s own judgment. If she is to meet the prevailing ethical standard, a surrogate decision-maker must somehow know how to translate the surrogated person’s values, beliefs, etc., into a decision that the latter would make for themselves. Of course, this knowledge, too, is often difficult to acquire. Even if it were easy to peer into an incapacitated and incommunicative person’s soul, it may not be easy to convert the knowledge discovered there into a decision with which the surrogated person would agree, if they could decide for themselves.
Third, a surrogate decision-maker might possess adequate knowledge of a surrogated person’s values, beliefs, etc., and of the particular decisions implicit in these values and beliefs, yet be incapable of implementing the implied decision in a way that realizes its apparent goal. For example, a medical surrogate might know that a surrogated patient’s relevant values imply that they would opt for a painless death under given circumstances, but be ignorant how to deliberately realize the surrogated patient’s peaceful shuffling off this mortal coil. Even if it were easy to both peer into a surrogated person’s soul and translate their beliefs, desires, values, etc., into a concrete decision or action, it need not be easy to act on that decision, to implement the implied action, so that its goals, whatever they might be, are ultimately realized.
That a surrogate decision-maker’s relevant knowledge is adequate is a precondition of ethical surrogate decision-making. A relevantly ignorant surrogate decision-maker cannot meet the standards of ethical surrogate decision-making, except, as it were, accidentally or spontaneously. The epistemic burdens of the prevailing ethical standards of surrogate decision-making can be quite heavy and may be, and perhaps, often are, difficult, if not impossible, for cognitively-limited surrogate decision-makers to surmount. The prevailing ethical standards of surrogate decision-making may make it difficult, if not impossible, to be an ethically good surrogate decision-maker.
Much has been written on the reliability of surrogates and their ability, or lack thereof, to make decisions that meet prevailing ethical standards.[3] That it can be epistemically challenging to decide as a surrogate decision-maker ostensibly ought to decide according to prevailing ethical standards is not a new point.[4]
What is novel about our research on the problem of policymaker ignorance and, more generally, on the epistemic difficulties of surrogate decision-making – for policymakers are just a particular kind of surrogate decision-maker; they make various decisions on behalf and ostensibly in the interests of their constituents – is the crucial, often determinative, role that epistemic burdens play in shaping the incentives of surrogate decision-makers to pursue either the goal of judging in accordance with the would-be judgments of the surrogated person or some other less epistemically burdensome goal, presumably less, if at all, consistent with the would-be judgments of the surrogated person.[5]
If a surrogate decision-maker recognizes both their relevant ignorance regarding the goal of properly substituting their judgment for the surrogated person’s judgment and their superior knowledge concerning some other goal, then, other things equal, they face a greater incentive to neglect the goal implied by the ethical standard and pursue the second, less proper, goal. In cases of extreme ignorance, a surrogate decision-maker might have no incentive whatsoever to conform to the ethical standard (and, perhaps, only incentives to decide unethically).
That epistemic burdens shape incentives is true for all decisions, whether personal or surrogate. This is the main implication of the thesis of the logical priority of the epistemic. However, given the obvious disconnect between the ultimate decision-maker and the person most directly affected by the decision, ignorance of relevant considerations is more likely in surrogate cases. Indeed, the divide between decider and decided-for makes the priority of the epistemic more starkly apparent in surrogate than in personal decision-making cases.
In short, the problem we are pointing to in our work on policymaker and, more generally, surrogate ignorance is not merely that surrogate decision-makers may be so ignorant that they cannot satisfy prevailing ethical standards. It is that they may be so ignorant – or, more exactly, that they may take themselves to be so ignorant – that they don’t even try.
So, you might ask, what is to be done?
As I’ve noted before, the only full-fledged solution to the specific problem of policymaker ignorance would be to make policymakers truly omniscient and omnipotent, to make them what they could never be, namely, all-knowing and all-powerful gods. Since this is impossible, we cannot solve, but can only hope to mitigate, the problem of policymaker ignorance.
We cannot make policymakers truly omniscient and omnipotent, but we might go some way toward making them functionally omniscient and omnipotent with regard to their specific policymaking purposes. In particular, we might either develop improved mechanisms for communicating relevant knowledge between policymakers and constituents or limit policymakers’ range of motion to the pursuit of goals with respect to which their knowledge is adequate. We might either try to provide political actors, policymakers and constituents alike, with the knowledge required to successfully fulfill their respective roles or simplify the requirements of good policymaking to make them more consistent with the cognitive limits of cognitively-limited policymakers. Neither of these methods would truly solve the problem of policymaker ignorance, but they might, either individually or in combination, constrain its capacity for wreaking havoc on the consequences of political decisions.
In principle and, perhaps in a few cases, also in practice, similar strategies might be possible for mitigating the more general problem of surrogate ignorance. We could never make human surrogates truly omniscient and omnipotent, but we might either improve their relevant knowledge to an adequate degree or simplify the ethical requirements of good surrogate decision-making to bring them into better accord with the typical epistemic limitations of surrogate decision-makers.
However, the problem with trying to improve the relevant knowledge of surrogate decision-makers is the very same incapacity, especially the incapacity of many surrogated persons in medical contexts to effectively communicate, that necessitates a surrogate decision-maker in the first place. In order to overcome their relevant epistemic burdens, surrogate decision-makers will often require the input of the surrogated person; yet, if the latter could communicate, the former would be unnecessary. So, the possibility of mitigating the problem of surrogate ignorance might be a non-starter in such cases.
This leaves the possibility of rejecting the substituted-judgment (and best-interests) standard in favor of another that better respects the cognitive limitations of human decision-makers. I have no idea what this new standard might be, but I do know that neglecting the implications of the problem of policymaker and, more generally, surrogate ignorance, as we have done for far too long, is no answer at all.
[1] The substituted-judgment standard is common in medical contexts. There is also a best-interests standard that holds a surrogate should decide according to what she takes to be the patient’s interests. Whether the appropriate criterion is substituted-judgment or best-interests has no bearing on the argument offered here.
[2] Recall that “knowledge” encompasses both propositional knowledge-that and non-propositional knowledge-how, both knowledge of theories and facts, and practical capacities, abilities, capabilities, powers, talents, and skills. To be ignorant is to lack either (or both) kind of knowledge.
[3] Berger J.T., DeRenzo E.G., and Schwartz J. (2008). Surrogate Decision Making: Reconciling Ethical Theory and Clinical Practice. Annals of Internal Medicine, 149 (1). DOI: 10.7326/0003-4819-149-1-200807010-00010.
Moorman S.M. and Carr D. (2008). Spouses’ Effectiveness as End-of-Life Health Care Surrogates: Accuracy, Uncertainty, and Errors of Overtreatment or Undertreatment. The Gerontologist, 48 (6): 811–819. DOI: 10.1093/geront/48.6.811.
Shalowitz D.I., Garrett-Mayer E., and Wendler D. (2006). The Accuracy of Surrogate Decision Makers. Archives of Internal Medicine, 166 (5): 493. DOI: 10.1001/archinte.166.5.493.
Suhl J., Simons P., Reedy T., et al. (1994). Myth of Substituted Judgment. Archives of Internal Medicine, 154 (1). DOI: 10.1001/archinte.1994.00420010122014.
[4] Brudney D. (2018). The Different Moral Bases of Patient and Surrogate Decision-Making. Hastings Center Report, 48 (1): 37–41. DOI: 10.1002/hast.809.
Buchanan A.E. and Brock D.W. (1989). Deciding for Others: The Ethics of Surrogate Decision Making. Studies in Philosophy and Health Policy. Cambridge University Press.
Pope T.M. (2012). Legal Fundamentals of Surrogate Decision Making. Chest, 141(4): 1074–1081. DOI: 10.1378/chest.11-2336.
[5] Kibbe and Ford (2016) come close to, but do not draw, the relevant conclusion.
See Kibbe B. and Ford P.J. (2016) What’s Knowledge Got to Do with It? Ethics, Epistemology, and Intractable Conflicts in the Medical Setting. The Journal of clinical ethics, 27 (4): 352–358.