CONCLUSION: THE CLAIMS OF DECISION-MAKING DEPTH
Man does not strive for pleasure; only the Englishman does.
1. Nietzsche, Twilight of the Idols, Maxim 12
The previous three chapters have argued that the foundation of the utility framework does not capture all of the essential elements of human decision making, particularly for those decisions that are infrequent or involve relatively important issues in our lives. Chapter 2 argued that the Principle of Optimality does not make sense in a world where decision-makers must take action in a stream of time with an uncertain future, or must forgo a substantial proportion of potential experiences. Chapter 3 argued that the principles of Commensurability and Sufficiency of Preferences implicitly rely upon a prior decision regarding the relative weights of various objectives, and it is in this prior decision where most of the interesting action occurs. Finally, Chapter 4 argued that the theme of direct pursuit underlying the utility framework is inconsistent with many desirable objectives that are not amenable to conscious attainment. Additional structure must be placed on the concept of the will to properly capture the role that conscious attention plays in reaching our goals.
In the introduction, I stressed the overwhelming complexity of human decision making – and I hope by now I have convinced the reader of the need to eliminate the drive toward parsimony in explaining it.
The model of decision making emerging from the previous chapters can perhaps be best highlighted by noting that human decision making is characterized by both depth and imperfections.1
Our decision-making depth is of two sorts: temporal and self- conscious.2 Our temporal depth allows us to make projections for the future and choose actions to try to bring such protensions to pass. We have the ability to visualize future events in the pluperfect tense, as already having happened 3.
Conscious depth allows us to both make decisions and know we are making decisions. Without this depth, our decisions would be mere actions, in much the same sense in which a heat-seeking missile does not decide, but does act. In its search for parsimony, the utility framework unfortunately makes no provisions for this depth. And it is this decision-making depth that distinguishes our decision making from that of animals.4
Along with this depth, human decision making is also characterized by imperfections. These imperfections arise from our inability to experience everything, the need to live through a stream of time, and the requirement to classify in order to process information from the external world.
The nature of these imperfections can be clearly seen if we contrast our situation with that of an omniscient being. Because such a being would be everywhere, everywhen, he would not have experiences in the sense we understand nor would he live in a stream of time. Because his thought processes would encompass particulars directly, he would not have to rely upon classifying objects into general categories. And because his thinking and being would not be distinct – his thoughts would simply be – the object of his reflection would completely coincide, eliminating the issue of self- consciousness. This omniscient creature, incidentally, is the God of Christian theology and is, of course, perfectly rational (a theme most systematically explored by Kant).
I assert that the decision-making situation in which we find ourselves may most accurately be described as being bounded rationality. Indeed, Simon’s original article [80] stressed the need to consider the conditions under which human beings actually choose; Simon then went on to examine more carefully the limits on our ability to calculate. I agree that we need to examine the conditions under which we actually choose, but I would argue that these conditions are precisely those examined in the previous three chapters: (i) we must classify objects in order to think and such classification requires a degree of non-deductive judgment; (ii) we are capable of self- reflection and some measure of choice over what we value; and (iii) our will is imperfect. That our rationality is bounded, therefore, distinguishes us from God.
It is interesting to note that the utility framework has borrowed from both models – animal and omniscient. The neglect of the need to classify corresponds to the abilities of the omniscient being, while the lack of self-consciousness corresponds to the decision making of animals.
The task for any investigation of human decision making, including this thesis, is to fit a decision-making framework into the narrow space between the limitations imposed by bounded rationality and the complexities implied by decision-making depth.
Having said all this, the obvious question is: Where do we go from here? Given that the utility framework does not capture many essential elements of our bounded rationality or our decision-making depth, what lessons are we to draw?
To explore these next steps, we need to spend some time developing the crucial distinction between the investigation of ourselves as objects of inquiry – that is, where human beings are both the subject and object of the inquiry and the exercise is one of self-interpretation – and the typical investigation of non-human objects. I assert that the conclusions drawn from the investigations of human decision making necessarily have consequences for the way we should decide (and I use “should” in the sense of how to ensure one lives one’s life in the best possible manner). That is, if in the course of our investigations we discover something about ourselves as decision-makers, this necessarily gives rise to claims regarding how we can rationally act. These claims arise as a consequence of the same self-consciousness that has animated much of this thesis. Put simply, if one discovers something in the world that applies to oneself, then one’s capacity for self-interpretation precludes this fact from being ignored when one acts or decides. Ideally, the set of claims generated in this way can be integrated into a strategy for ensuring that one’s life goes as well as possible.
Thus, for example, if an individual discovers that his aggressive behavior can be traced to incidents of childhood abuse then this knowledge will immediately affect his understanding of himself and his actions, and may have some implications for future behavior. Similarly, someone who discovers that he has been systematically underestimating the negative impact on his family life of his decision to work long hours must also recognize the implications that stem from how he allocates his time among multiple commitments. A familiar example might be the issue of regret in decision theory. If decision theorists teach us that we generally experience regret as a result of a good ex ante decision turning out poorly ex post, then this should have a bearing on how we make decisions. Once one recognizes regret, then one either avoids it by figuring it into utility calculations in advance or, better still, realizes that one should not feel regret and thereby gradually learns not to experience it. In all these examples, the discovery of additional information concerning the conditions under which decisions are made gives rise to claims that must be addressed by the rational decision maker.
This generation of claims regarding our self-interpretations and behavior can also occur on a much broader scale. Kant’s notions concerning human equality are now firmly ingrained in the way we understand ourselves, and how we treat others, both legally and socially. It is of no consequence whether Kant is actually right – indeed, this question is unanswerable. It is rather that in accepting his arguments concerning equality into our self-interpretations, we must also take the consequences of this doctrine seriously when we make laws or interact socially. Rationally, we cannot embrace Kant’s principle of never treating others as means while simultaneously bidding for slaves in the Confederate South. Freud’s theories on the subconscious are a second example of a self- interpretation that is deeply embedded within our daily lives. Accepting Freud’s assertions regarding the existence of a super-ego necessarily means accepting a different way of viewing oneself, one that must include Freudian concepts.
Whether or not we agree with Hegel’s theories of the ultimate rationality of history, we must agree that man is an evolving animal in that self-interpretations and concrete actions, rightly or wrongly, are changed as he learns more about himself.5
These claims do not necessarily arise when the object of the investigation is not oneself – that is, in those investigations where the results do not contribute to one’s self-understanding or self- interpretation. Consider, for example, an investigation that determines the pattern of coastal erosion resulting from prevailing winds off the Great Lakes. Does this discovery engender a claim for how one acts and decides in the world? Not necessarily. It may, of course, if one were considering building a cottage on a spit of land likely to be swept away by erosion but, in general, such investigations have no direct claim on our attention. Their relevance is contingent on the particular discovery having a relationship to plans or objectives we may have in the physical world. And it would not be irrational to ignore the results within one’s personal life if they were not felt to be relevant. But if the object of inquiry were oneself, it is difficult to imagine that subsequent discoveries would not be sufficiently relevant to generate such claims.
Furthermore, claim recognition must be a cornerstone of any theory of substantive rationality. Advocates of the utility framework have generally seen our pretensions to rationality as being founded on our ability to perform means-end calculations with given preferences and subjective probabilities. This theme has recurred repeatedly in this thesis. However, this can only be part of what constitutes rationality for beings who possess the capacity for self- interpretation. To repeatedly ignore known claims is to act irrationally. Habermas [24:21] captures the spirit of this observation in developing his view of rationality:
Anyone who systematically deceives himself about himself is acting irrationally. But one who is capable of letting himself be enlightened about his irrationality possesses not only the rationality of a subject who is competent to judge facts and who acts in a purposive-rational way … he also possesses the power to behave reflectively in relation to his subjectivity and to see through the irrational limits to which his cognitive, moral-practical, and aesthetic- practical expressions are limited.
Rationality can only partly consist of the instrumental ability so essential to the utility framework and other rationalist or reductionist models of human behavior. Our capacity for self- interpretation implies that rationality must be further extended to include the recognition of these legitimate claims arising from self knowledge. One must be prepared to be enlightened about oneself, and be prepared to act upon this enlightenment.
It follows, therefore, that if we accept that rational claims are generated by understanding how we make decisions, then the arguments of the preceding three chapters must give rise to such claims regarding how we can coherently think about our actions and choices. Indeed, a corollary to the principle that self-referential investigations have implications for the way we should understand our own actions and for how we should act is that worthwhile insights must result in some additional claims on the individual. A psychologist, for instance, could not conclude that his work was worthwhile unless it had also generated claims of the sort I am describing. If he had not sufficiently promoted the understanding of the reasons behind individual choice and self-interpretation so that at least some individuals were confronted with these claims, then it is difficult to see how his work could be a true contribution.
The next step, then, is to investigate more systematically the claims arising from the arguments of the previous chapters. Let me say at the outset that I will describe these claims only in broad strokes. This thesis merely lays the groundwork for a more thorough analysis of the issues and generation of claims – an analysis that would necessarily include both further analytical thinking and perhaps some empirical work.
The method I will use will be to contrast the claims that result from the utility framework – for that too must give rise to claims in precisely the same manner as any theory of human action – with those that arise from the previous three chapters. First to be considered are those claims resulting from the considerations raised in Chapter 2.
5.1 The Claims of Judgment
Let us begin by exploring the claims that arise if we accept the utility framework as the appropriate way to view human decision making. Under this model, the most pressing problem facing the decision maker is that of applying a certain technical or instrumental rationality to secure what he desires. Claims arise from this model of human decision making because certain choices are ruled irrational since they do not correspond with these rules of technical rationality. The expected utility theorem of von Neumann and Morgenstern [49] is a striking example of one of the claims that arises from the utility framework. Given the structure of human decision making implicit in the theorem, an individual is choosing rationally when he chooses such that his expected utility is maximized. Preferences are known and fixed, and probabilities are subjective in this model, as described in Chapter 2. By asserting that deliberate pursuit of utility maximization is rational, von Neumann and Morgenstern generate claims resting on their theory of economic behavior [49:8-9]:
We shall . . . assume that the aim of all participants in the economic system, consumers or entrepreneurs, is money, or equivalently a single monetary commodity . . . The individual who attempts to obtain these respective maxima is also said to act “rationally.”
Given this view of rationality, the demand on the self-conscious agent is that of best attaining these maxima. The rational individual is compelled to choose the plan of action that best secures what he wants, given his subjective view of what is likely to happen in the world.6 Decision theory is one of the tools that this actor can use when the decision situation is particularly complex.7
It would indeed be comforting if this model of conditions under which individuals must actually choose was accurate. In this case, the claims emanating from it – a sort of strict technical rationality resident in specialist planners and managers – would be decisive.
Unfortunately, as Chapter 2 argues, this view of practical epistemology does not fit with the conditions under which we do make choices. Judgment – which requires the construction of categories to manage the inherent uncertainty in the world – is of much greater importance for the making of good decisions than the ability to manipulate calculations after the relevant categories and probabilities have been chosen.
At the most general level, then, the claim that arises from this observation is that decision-makers should concentrate more on developing good judgment than on perfecting calculating abilities. Such a claim, however, is of no practical value. Certainly, we should improve our judgment – but saying so will not make it so. Nor is there an easy answer to how we do so.
We can discern the nature of this difficulty by comparing the reasons for which we deem correct a calculated decision versus those for which we deem correct a decision requiring a high degree of judgment.
In the former case, the decision is easily checked by ensuring that the calculations have been done correctly – How much flour do we need to bake 12 dozen cookies? Multiply the recipe for one dozen by 12, And we can always determine whether we have the right answer.
With decisions requiring a high degree of judgment, however, such an approach is useless. There is simply no independent check on a contemplated decision – Should Chrysler allocate its research resources based on the premise that light trucks and vans will capture 50 per cent of the domestic vehicle market by 1994? Who knows? In cases such as this, in lieu of examining the decision itself and determining whether the choosing of categories and the employing of analogical reasoning have been done adequately, we typically evaluate the decision-maker and his track record.
Indeed, it is this principle that governs success or failure in many professions. Businessmen are promoted because their excellent track records are thought to reflect able judgment. Consultants, having little tangible to offer clients, are hired because they have displayed good judgment in the past. Leaders are chosen because their past successes demonstrate their good judgment and their decisions therefore will carry extra weight. Conversely, unproven individuals are far less likely to have their proposals accepted – even if these proposals are identical to those* offered by people with a known and respected reputation. Aristotle makes much the same point in describing his epistemology of ethics. Whereas determinate rules can be given for all other types of knowledge, that from “prudence” (which corresponds to our judgment) seems to stem only from experience [2:Book Z].
This insight provides a glimpse of just how difficult it is to judge judgment. If we had a better understanding of its requirements, we would not need to place the same reliance upon reputation for we could then examine the decision itself – much as we can with calculated decisions.
This complex issue is brought into relief if we consider attempts at codifying the professional teaching of judgment. The problem arises because many of the skills that we desire the next generation’s practitioners to have fall under the category of judgment rather than analytic ability. Thus, despite repeated attempts to neatly delineate the process doctors, lawyers, engineers, and other professionals follow in doing their jobs – delineation that would be exceptionally convenient for teaching these abilities – the required skills are remarkably resistant to codification. Schon [72,73] does an excellent job of describing the sense of uneasiness that accompanies the profession’s attempt to teach the art of judgment. This art does not sit well with what Schon [73:6] terms the “dominant model of human knowledge”: professional practice as an “exercise of technical rationality, that is, the application of research-based knowledge to the solution of problems of instrumental choice.” The difficulty here is not that competent professionals do not exercise good judgment; rather it is that no one has discovered a systematic way of transmitting judgment through teaching. As a result, professional schools concern themselves instead with teaching the purely mechanical parts of the profession.
Having said all this, the issue still remains of what claims – concrete steps that arise through the considerations of judgment – the concept of judgment imposes on decision-makers. I think there are some, and I will identify two sorts of them. The first is based on the need to take the difficulties of exercising judgment seriously, and the second recognizes the role options must play when individuals are faced with the kind of uncertainty described in Chapter 2.
Scenario analysis, a research program investigating systematic ways of improving judgment and analogical reasoning in the sphere of business decision making, strikes me as being on the right track with the first claim. It begins with the premise that simple decision theory tools (calculation tools) only solve the simple problems. If we take our assumptions as given, it is relatively straightforward to compute the optimal decision (recall that money is our single yardstick, so the problem of commensurability disappears in business). The key to good judgment, then, lies in choosing the right framework or structure to describe the problem. Framework and structure correspond to the idea of category presented in Chapter 2.8. I should add that scenario analysis is not an attempt to bring an analytic technique to bear in an inappropriate situation. The element of personal judgment – choosing the structure, fixing the framework, or defining the relevant categories and characteristics – is not eliminated. The advantage in scenario analysis lies solely in bringing the already extant judgment or intuition to a problem in a slightly more systematic way. The ultimate decision-maker can make effective use of analytic techniques but, as always, must in the end choose for himself when judgment is required.
Scenario analyzing is most useful when we are in “uncharted waters” – as the titles of Pierre Wack’s [92,93] seminal papers indicate. When business is as usual, calculation and deductive reasoning may suffice; when genuine uncertainty prevails, scenario analysis gives some structure to the application of judgment.
A kind of personal scenario analysis – periodically checked to explore how proposed scenarios and actual events or preferences diverged – would go a long way toward systemizing and improving human judgment. By bringing judgment to bear in this formal way, we force recognition of the inherent uncertainties in life and confront the irreversibility and uniqueness of our decision situation. Finally, scenario analysis would move the locus of attention away from the purely deductive elements of decision theory to the aspect of judgment that is so much more important for making good decisions.
Of greater significance, however is the second sort of claim arising from Chapter 2 – that of the role personal options should play in any attempt to make best possible choice, especially in those situations in which genuine uncertainty prevails.
Traditional models of choice based on the utility framework have no room for the importance of real choice options. These models are essentially metastatic [90] – individuals choose once for all time. Even extensions that are not metastatic require individuals to leave their contingent planning unchanged once the clock has started ticking [44]. The model here is one of the decision-maker examining the future possibilities, subjective probabilities, etc., and choosing an action plan which maximizes expected utility given the individual’s subjective probability – a model is similar to the discounted cashflow calculations used to perform project evaluation in business. Of course, the decision-maker recalculates his optimal decisions as uncertainty is resolved. This, however, is not equivalent to building options in from the start in the decision process.
In conditions of true uncertainty, and in which the decision is unlikely to be repeated, the value to the individual really comes from the existence of real options – the opportunity to follow as many different paths for as long as possible.
It is only now that academics are discovering both the value of and the role played by real operating options in business decision making. The best managers do not rely upon their discounted cash flow forecasts – which are typically examples of business uncertainty dressed up in the semblance of certainty or riskiness – for they recognize that there are always discrete exit points, possibly with some salvage value. Indeed, management is paid precisely because they have the ability to change plans as uncertainty is resolved. What appears to be a poor project in the metastatic discounted cash flow view looks very attractive when the real operating options are taken into account.
Given the inherent difficulties of judgment, the claim on the individual is to postpone as many unique or irreversible decisions as possible. There is no doubt that individuals actually choose in this way when confronted with significant decisions with great uncertainty. Thus, I may decide to live with someone before going through with the marriage, or structure my college work so that I can enter either medical school or business, or choose a summer job in which I spend half my time practising law in Los Angeles and the other half practising in New York. These strategies recognize that disengaging from prior choices can be difficult and careful planning is required to build in real operating options.
Putting some structure on the role and value these options play is the route that decision theorists should pursue if they wish to help people make better decisions. The relevant question to ask the decision-maker is not his subjective probability of becoming a partner in his firm, elicited by the hypothetical wagers. Rather, the question is what options will he have if he finds his aptitude or inclinations are not suitable for the profession. This is the sensible way of making decisions, and it is one of the key claims arising from the arguments of Chapter 2.
Moving in this direction would bring us that much closer to thinking about the conditions under which we actually choose. And careful thinking about options, while not amenable to calculable models, can probably be captured in analytic models similar to those in analytic philosophy.
I have two final comments before addressing the claims of autonomy and appropriateness. The first is that a major reason for not studying this operation of judgment in sufficient detail is probably because much of the action here happens in the unconscious.10 If this hypothesis were indeed true, then perhaps economists and decision theorists can hide behind the academic division of labor in explaining their reluctance to address the distinction I have drawn. This would still be an admission, however, that further insight into individual decision making is not going to come from refinements in the utility framework. The important action is going to occur elsewhere.
The second is that this mysterious process of judgment, which plays the key role in decision making, should not be viewed with derision because it blocks the scientific program of Liebniz, Carnap, and others. Life retains its interest solely because we must make uncertain judgments regarding an uncertain future with categories of thought that are open and ambiguous. Were this not the case, we could turn the world over to Liebniz1 calculators – and the world would be the poorer for it.
5.2 The Claims of Autonomy
In this section, I address the claims that arise from the capacity of self-conscious agents to deliberate over what is to be valued. Because its parsimony does not allow the utility framework to fully reflect the issues discussed in Chapter 3, it gives rise to no particular claims. The utility framework simply does not have sufficient structure within it to consider questions of autonomy.
By developing the capacity for autonomy, the individual gains a certain measure of control over his preferences, as well as a critical awareness of the existence of competing preferences. In this way, preferences become truly his, as he regulates the process of preference generation and only admits preferences once they meet his criteria.
One claim that I do not think can be made, at least not yet, is that autonomy is a desirable goal and we should all strive to achieve it. The argument in favor of autonomy is at best ambiguous. It is not clear that utility – to use the economist’s term – is increased when we choose. Those who blindly follow preferences absorbed from their cultural milieu may be as happy, if not happier, than those who are confronted with the necessity to choose. This issue quickly gets us into Mill’s “satisfied pigs” argument or Dostoevsky’s inquisitor.
In this sense, then, autonomy is very different from judgment. No one can sensibly dispute that developing good judgment is a desirable goal. And once we recognize the essential role judgment must play in our lives, the claim of developing good judgment is decisive.
Now there are of course those who do believe that possessing autonomy is a desirable goal (and I count myself among this group), but the argument here is generally based not on the fact that our lives will be more satisfying11 but on a semi-naturalistic argument that it is our nature to live our lives self-consciously rather than to sink to the purely animal level of following our desires blindly. Indeed, Frankfurt’s disparaging term “wanton,” used to describe those who fail to exercise their capacity to evaluate their preferences, captures this attitude that one is not completely human if one does not engage in this exercise of rational deliberation. Nonetheless, the disparagement itself is no proof of a claim that the rational individual must follow. Without first providing a convincing argument, we can hardly describe as irrational the decision-maker who blindly follows his desires without exercising capacity for agency.
Having said all this, and despite the fact that there is no evidence that an individual is better off if he chooses autonomously, there is a claim in those cases where the individual becomes aware of his capacity to affect his preferences – the claim of responsibility. Once the capacity for autonomy is recognized by a self-conscious agent, he cannot turn his back on its force. An individual who admits his potential to deliberate over possible desirable ends must also admit the necessity to do so. There is no returning from the state of autonomy to non-autonomy – there can only be an autonomous decision to abdicate responsibility for choosing one’s preferences.
In this case, one is deliberately making the choice not to choose, and that too is an affirmative choice. The unfortunate fact that it is impossible to return to the naivete of non-autonomy once the potential for autonomy has been recognized has deep roots in Christian thought and Western society in general. A lack of self- consciousness has always been interpreted as innocence in our society, as witnessed by our attitude toward the un-self-consciousness of children or by the biblical tale of Adam, Eve, and the forbidden fruit of knowledge.
This notion of responsibility for choosing one’s own preferences perhaps finds its best expression in the discussion of evaluation put forward by Taylor [85:39]:
Responsibility falls to us in the sense that it is always possible that fresh insight might alter my evaluations and hence even myself for the better. So that within the limits of my capacity to change myself by fresh insight … I am responsible . . . for my evaluations.
Taylor’s treatment of this issue differs from that of Nietzsche and Sartre, in that it does not rest so much on the suprahuman ability for radical choice discussed in Chapters 3 and 4. For Taylor, this responsibility is simply the recognition that one does have a measure of control over one’s evaluations (or preferences), and that examination of one’s preferences must somehow be rooted in one’s own experience, one’s own narrative.
The utility framework has never come to grips with this notion of responsibility. The title of the classic Stigler and Becker [83]article, “De gustibus non est disputandem,” illustrates well the predominant idea within the economics profession – tastes are beyond the realm of debates concerning choice. Arrow [3:17], too, makes a similar claim: Rationality, after all, has to do with means and ends and their relation. It does not specify what the ends are.
I hope that the arguments of Chapter 3 and this chapter have convinced the reader that responsibility for evaluations must form an integral part of any framework for examining choice and of any set of personal evaluations for an individual who recognizes his capacity for this evaluation. Further, these evaluations cannot be totally divorced from substantive rationality.
5.3 The Claims of Appropriateness
The arguments raised in Chapter 4 also lead to claims, but ones that may not be easily realizable in practice. In developing appropriateness, one gains a certain degree of self-mastery and ceases to channel energy in seeking goals that are fundamentally unreachable through direct pursuit.
The exhortation, if we go back to the quote of Mill from his Autobiography, is to enjoy life in passing. From the viewpoint of phenomenology, we might say that the goal is to prevent the eye of conscious attention from settling for too long in one place, because this attention is destructive. I do not think there should be much dispute that appropriateness is a desirable quality for an individual to possess. Our lives would all be improved if we could devote the required degree of conscious attention to our projects.
I think the notion here is best captured by Nietzsche’s arguments concerning willing and second nature. Schacht [69:280] describes Nietzsche’s view here:
. . .”becoming-instinctive” for Nietzsche . . . may also have the significance of an advance beyond [normal] consciousness, to a new and “higher naturalness” . . .
Having attained such a state, one is no longer in a position to have to think about one’s actions and objectives. Activity is still goal- directed, but without the anxiety discussed at length in Chapter 4. When Nietzsche [52:440] asserts that “one who acts perfectly only when one acts instinctively,” the notion is that one’s engagement in the activity and its objectives is completely natural and, indeed, instinctive. But this perfect action does not arise from any natural instinct; rather, the objective and means have been chosen by the decision-maker. It is the execution of the activity that has reached the perfection implied by instinct. In this sense, the activity has become second nature, a nature which has been chosen and developed. Upon reaching this ideal, the decision-maker has advanced beyond the difficulties raised in Chapter 4.
Nonetheless, as a claim, the difficulty is that it is impossible to follow consciously or deliberately. Indeed, that is the essential argument of Chapter 4: the notion of an optimal degree of conscious attention actively chosen is incoherent, because there is no mechanism for the self-conscious decision-maker to ensure that he only applies the optimal amount. The degree of conscious attention cannot itself be a conscious decision.
Again, indirect strategies are the only salvation, but only in those cases where one genuinely fools oneself or forgets the reason lying behind the strategy. Nietzsche, despite bringing our attention to the possibility of this advance in the efficacy of our efforts to reach our goals, is of no practical help. Within his framework, only certain individuals possess this capability, such men being labeled Overman (ubermensch); the great bulk of mankind is deemed unable to achieve this higher state. There is no advice on how to be one of these few – it is simply something that one must sort out for oneself.
We can conclude that, despite the existence claims arising from the considerations of Chapter 4 (that of achieving the appropriate combination of conscious and unconscious willing), it is not immediately clear how the rational actor can go about fulfilling these claims.
5.4 Life’s Strategy
I said above that the collection of claims arising from the examination of our situation will result in a strategy (or strategies) towards life. The articulation of strategies in life – both those suggested by considerations of these claims and those actually used by individuals – strikes me as the logical next step in this endeavor. This articulation of a life strategy would include:
1. An approach to resolving uncertainty – search methods, options generations, information gathering, decision triaging, etc.
2. A method for managing preferences – how relative weights are decided, situations in which the capacity for agency is engaged, etc.
3. Examining the degree of conscious intentions underlying the pursuit of desirable objectives (possibly including indirect methods for diverting conscious attention), one’s general attitude toward life, etc.
The empirical questions here concern identifying which life strategies are most successful (the measurement of success, unfortunately, will be fraught with all sorts of difficulties), those strategies actually employed by individuals, and the possibilities and methods for moving from one strategy to another.
One important result of the articulation of actual life strategies would be the extensiveness of awareness of the issues raised in this thesis among the general population. Critics of this essay might hazard that Chapters 3 and 4, concerning the capacity to choose weightings and the difficulties with deliberately pursuing some desirable objectives, only apply to a tiny minority of the population. These critics might also assert that the utility framework is the most appropriate approach to human behavior for the vast majority who are unaffected by the issues raised in this thesis.
Of course, such critics must be careful. Asserting that most individuals are unconcerned with these issues would not necessarily imply that this ignorance is the optimal strategy for life – that is, Mill might be right about his satisfied pigs. It might very well turn out that many individuals do not bother with these complex issues and are much worse off for it.
The core issue, of course, is an empirical one and will only be settled by examining the strategies individuals actually follow. My hypothesis going into such an empirical investigation is that many individuals use sub-optimal strategies – why should individuals be any different from corporations? Indeed, the deviations from an optimal strategy are liable to be even more severe in individuals, as there is little competitive pressure pushing individuals to approach life in the most efficacoius manner. The strongest forms of descriptive theories based on the utility framework – Becker, for example – would be unable to even frame the issue of sub-optimal strategies.
Moreover, I would not be surprised by the following specific findings: (i) most people act as if events and preferences are more certain than, in fact, they are and do not build in appropriate options; (ii) many could increase happiness by weighing preferences more carefully; and (iii) a substantial proportion of individuals pursue desirable goals in an overly deliberate fashion.
If it is discovered that following poor life strategies is widespread, this too would generate claims of the type described in this chapter. And do recall that the rational individual cannot ignore the power of these claims. If, for example, it was determined that many individuals could increase overall welfare by paying more attention to the choice of weightings, then this knowledge will have implications for how much effort should be devoted to choosing preferences. If it turned out that most lawyers do not enjoy their profession – though earlier they consciously chose it – this fact should have ramifications for how one structures the exit options from a law career.
In building these claims, I hope that the proposed empirical investigation could take a step toward a practical and substantive rationality – a rationality that is sorely missing from debate grounded in the utility framework.
5.5 Conclusion
In the introduction, I stressed the need to take account of the conditions under which we actually choose when addressing questions of human choice. In this thesis, I have attempted to trace through some of these conditions and have contrasted these conclusions with the assumptions underlying the utility framework. The key points I have made about this framework are:
(I) The utility framework is the foundation that underpins microeconomics, ethical utilitarianism, and decision theory.
(II) The utility framework is attractive to social scientists and ethicists because it provides a rational mechanism for singling out a particular action or decision from the feasible set.
(III) The utility framework does not take adequate account of the inherent uncertainty decision-makers face, nor of the necessity to construct a conceptual model of the world.
(IV) There is no room within the utility framework to capture our capacity to evaluate and choose our preferences.
(V) The utility framework does not distinguish between different ways of pursuing desirable goals, nor does it address the role intentions must play in any theory of decision making.
(VI) The utility framework’s most pressing difficulties are to do with decisions that affect the most critical aspects of our lives.
In addition to dissecting these conceptual difficulties with the utility framework, I’ve tried to sketch out some of the ramifications of a more complex decision-making framework. These thoughts culminate in the claims described in this conclusion.
There are obviously many issues that have been left unaddressed by this thesis. This is inevitable in such an endeavor where one is trying to address the fundamental issues of choice and rationality. Moreover, despite the seeming futility of continuously discussing questions that have no definitive answers or closure, debate is still essential. It should not and cannot cease. It is the mechanism through which we grope for a more complete understanding of the circumstances in which we choose. In the process, we also develop a view of rationality, and greater self-understanding. I hope that I have made a contribution to this on-going debate.
ENDNOTES TO THE CONCLUSION
1. I hesitate to use the term “imperfections” here, as it suggests that “perfection” is somehow attainable. And perfection would necessarily correspond with the type of deductive reasoning that is at the core of calculation as described in Chapter 2. Our decision making is only imperfect when contrasted with that of the Christian God, an entity whose perfection is based on the existence of a unique rationality.
Nevertheless, given our starting point – Western and Christian thought – the term “imperfections” perhaps best conveys the particular boundedness of human decision making that this thesis has explored. In no way do I wish to suggest that perfection is attainable, desirable, or coherent as a goal of our decision-making.
2. The considerations arising from this decision-making depth are not equally applicable to all classes of decisions. We can triage preferences into the following categories:
(I) Preferences we simply have, where the deeper issues of life are not involved. With such preferences, there is little point to consciously choosing appropriate weights nor is utility much affected by whether our will is conscious or unconscious. (II) Preferences over which we have some control, and which may have the type of qualitative distinctions suggested in Chapter 3. In such cases, our capacity for agency is activated. (III) Preferences that touch the deeper values in life, but where too much conscious willing may actually lead to a diminution in satisfaction.
For the first category, our decision-making depth plays a limited role and the approach suggested by the utility framework is more than adequate. In these cases, our preferences are concerned with objects that do not relate to the deeper aspects of life, the difficulties with projecting unexperienced utilities and future events are usually not overwhelming, and direct pursuit of these desirable goals does not impede our ability to attain them. We should all be able to choose between salmon and chicken without paying much attention to our decision-making depth. Similarly, if I know the weather forecast, and my utility payoffs are dependent upon camping/staying home and rain/shine, I should be able to make this choice with the tools that decision theory (based on the utility framework) can muster.
The considerations of depth, then, have much more to do with the second and third category of preferences. These preferences generally involve issues that relate to self-definition and hence demand autonomy, or are not attainable through the typical goal- project-action sequence upon which the utility framework relies.
3. Readers interested in the phenomenological depths of this “visualizing actions as already having happened” are referred to Schutz [74: Chapter 2], Schutz also reviews Husserl’s important work in this area here.
4. It is interesting to observe that the capacity to even pose questions concerning decision making is itself evidence of the two types of depth. To ask how choice is possible and how we should characterize individual decision making is to implicitly acknowledge that we possess both self-consciousness and the ability to form the necessary protensions. Descartes’ cogito immediately comes to mind.
5. On a more mundane level, the plethora of self-help books available to increase one’s self-understanding points to the powerful sway of the claims generated in this way. By understanding the source of our motivations or actions fully, we gain a certain self-mastery; hence, the powerful appeal. For example, one can choose books that explain why we do not easily accept death, why women incorrectly blame themselves for becoming victims, and why individuals are often insufficiently assertive.
6. The consternation over the Allais and Ellsberg paradoxes (see, for example, [71]) illustrates the nature of the claim within the expected utility framework. Because rational individuals should choose a certain way – within the framework, that is – the empirical fact that many do not pushes us to tack on a theory of regret (or some other fix) to keep the model coherent.
7. Ryle [64:Chapter II] does an excellent job of tracing the source of this impulse to reduce all knowledge to a set of calculable procedures. His crucial distinction between “knowing how” and “knowing that” is relevant here. According to Ryle, there is a long history in Western thought attempting to reduce practical knowledge to a species of rules being followed. Demonstrating that such rules exist for all activities would be sufficient to show that “knowing how” can always be reduced to “knowing that.” The ability to throw a baseball, if this project were successful, could be transmitted from person to person simply by explaining baseball-throwing rules to the pupil. Once one has the rules, the correct procedure should follow. In this way, all knowledge is reduced to “knowing that,” and being able to name all the Balkan countries and hit a squash ball are, at bottom, the same kind of thing. Ryle correctly rejects this view, arguing that “knowing how” must precede the formulation of the rules, and that it is the practice that finally gives rise to the theory.
8. References for scenario analysis include the classic articles by Pierre Wack [92,93] as well Porter [59:Chapter 13].
9. A concrete problem related to the issue of judgment and scenario analysis concerns the role that so-called holistic reasoning might play.
Consider an approach to judgment that begins by breaking down the problem into pieces and finding an analogy for them. Thus, in asking the question: Is there intelligent life out there who is sending signals we could detect?, the first issue might be the probability of life at all. The trick is now finding an analogy for this single piece rather than the overall system. The initial classification is itself an exercise in judgment, though the mere fact of enumeration should help to improve judgment here. This last point is well illustrated by Howard Raiffa’s example of using decision theory to investigate why a car might not start. After teaching the lesson, his own car wouldn’t start. It turned out that he was in the wrong car, a possibility not enumerated in the class’s decision tree.
There is, however, a danger in this reductionist approach, good judgment may not be amenable to compartmentalization. The following situation illustrates the nature of the issue. Imagine one has hired an experienced consultant to evaluate the desirability of embarking on a new business venture. The consultant studies the problem and renders a verdict. Now imagine the consultant is brought in and grilled on how he reached his answer. His reasoning is painstakingly examined piece by piece. Imagine, finally, the consultant changes his verdict at the end of this session. The question is, which route is likely to be more profitable – following the original suggestion or the new, critically examined suggestion? To put it another way, has something been lost by the piece-by-piece examination? Should the consultant’s judgment be trusted because it encompasses some type of holistic reasoning? This question can only be settled by careful empirical work.
10. Knight [38], for example, places the activity of judgment somewhere in the netherworld of the unconscious.
11. Indeed, this issue raises some rather intriguing questions regarding Mill’s higher and lower pleasures. It may be that the man who has tasted both non-autonomous and autonomous preferences prefers non-autonomy, but how does he go back? As with the issues raised in Chapter 4, one cannot autonomously decide to be non-autonomous.
BIBLIOGRAPHY
[1] Ainslie, G. 1985. Beyond microeconomics: Conflict among interests in a multiple self as a determinant of value. In J. Elster (ed.) The multiple self, pp. 133-175. Cambridge: Cambridge University Press.
[2] Aristotle, 1984. Nicomachean ethics. Translated by Hippocrates G. Apostle. Grinnell, Iowa: The Peripatetic Press.
[3] Arrow, K. 1974. The limits of organization. New York: Norton.
[4] Becker, G. 1977. The economic approach to human behavior. Chicago: University of Chicago Press.
[5] Bell, D., Bradley, S., and Haimo, V. 1986. Scenario analysis. Harvard University mimeo.
[6] Bentham, J. 1963. An introduction to the principles of morals and legislation. New York: Harper.
[7] Berlin, I. 1969. Four essays on liberty, London: Oxford University Press.
[8] Berlin, I. 1980. Against the current: Essays in the history of ideas, New York: Viking Press.
[9] Camus, A. 1958, Caligula. Paris: Gallimard,
[10] Carnap, R. 1950. Logical foundations of probability. Chicago: University of Chicago Press.
[11] Connolly, W. 1983. The terms of political discourse, Oxford: Martin Robertson.
[12] Cyert, R. and DeGroot, M. 1975. Adaptive utility. In R. Day and T. Grove (eds.) Adaptive economic models, pp.223-46. New York: Academic Press.
[13] Deaton, A. and Muellbauer, J. 1980. Economics & consumer behavior. Cambridge: Cambridge University Press.
[14] de Villiers, P. and Herrnstein, R. 1977. Toward a law of response strength. Psychological bulletin 84, pp. 216-30.
[15] Dostoevsky, F. 1981. The brothers Karamazov. Translated by A. MacAndrew. New York: Bantam.
[16] Edgeworth, F.Y. 1881. Mathematical psychics. London: Kelley.
[17] Elster, J. 1979. Ulysses and the sirens: Studies in rationality and irrationality. Cambridge: Cambridge University Press.
[18] Elster, J. 1983. Sour grapes: Studies in the subversion of rationality. Cambridge: Cambridge University Press.
[19] Elster, J.(ed.) 1986. The multiple self. Cambridge: Cambridge University Press.
[20] Farber, L. 1976. Lying, despair, jealousy, envy, sex, suicide, drugs, and the good life. New York: Basic Books.
[21] Frankfurt, H. 1971. Freedom of the will and the concept of a person. Journal of philosophy 68, pp. 5-20.
[22] Friedman, M, 1953. The methodology of positive economics. In M. Friedman Essays in positive economics, pp. 3-43. Chicago: University of Chicago Press.
[23] Gallwey, W.T. 1979. The Inner Game of Tennis. New York: Bantam.
[24] Habermas, J. 1981. Reason and rationalization in society. Vol.1 of The theory of communicative action. Translated by T. McCarthy. Boston: Beacon Press.
[25] Hacking, I. 1975. The emergence of probability. Cambridge: Cambridge University Press,
[26] Hare, R. 1963. Freedom and reason. Oxford: Oxford University Press.
[27] Harsanyi, J. 1982. Morality and the theory of rational behavior. In A. Sen and B. Williams (eds.) Utilitarianism and beyond, pp. 39-62. Cambridge: Maison des Sciences de 1’Homme and Cambridge University Press.
[28] Harsanyi, J. 1986. Advances in understanding rational behavior. In J. Elster (ed.) Rational choice, pp. 82-107. New York: New York University Press.
[29] Herrnstein, R. 1986. A behavioral alternative to utility maximization. Harvard University mimeo,
[30] Hirschman, A. 1986. Rival views of market society and other recent essays. New York: Viking Press.
[31] Hume, D. 1962 [1739]. A treatise of human nature. Book One. Edited with an introduction by D.G.C. MacNabb. Glasgow: Fontana/Collins.
[32] Jeffreys, H. 1983. Theory of probability. Oxford: Oxford University Press.
[33] Johnson, S., Kotlikoff, L, and Samuelson, W. 1986. Empirical life-cycle models. Harvard University mimeo.
[34] Kahneman, D. , Slovic, P. and Tversky, A. (editors) 1982. Judgment under uncertainty: Heuristics and biases. Cambridge: Cambridge University Press.
[35] Kant, I. 1933. Critique of pure reason. Translated by N.K. Smith. London: The MacMillan Press.
[36] Kant, I. 1948. The moral law: Kant’s groundwork for the metaphysics of morals. Edited and translated by Jf.J. Paton. London: Hutchinson.
[37] Keynes, J.M. 1921. A treatise on probability. London: AMS.
[38] Knight, F. 1921. Risk, uncertainty, and profit. New York: Kelley.
[39] Kuhn, T. 1970. The structure of scientific revolution. London: The University of Chicago Press.
[40] Lancaster, K. 1971. Consumer demand. New York: Columbia University Press.
[41] Luce, R.D. and Raiffa, H. 1957. Games and decisions: Introduction and critical survey, New York: John Wiley & Sons.
[42] MacIntyre, A. 1981. After virtue: A study in moral theory. London: Gerald Duckworth.
[43] March, G. 1986. Bounded rationality. In J. Elster (ed.) Rational choice, pp. 142-70. New York: New York University Press.
[44] Marschak, T. 1979. On the study of taste changing policies. American economic association proceedings 68, pp. 386-391.
[45] Mill, J.S. 1924. Autobiography. Reprint. New York: Columbia University Press.
[46] Mill, J.S. 1978. On liberty. Indianapolis: Hackett Publishing.
[47] Mill, J.S. 1979. Utilitarianism. Indianapolis: Hackett Publishing.
[48] Nelson, R. and Winter, S. 1982. An evolutionary theory of economic change. Cambridge, Massachusetts: The Belknap Press of Harvard University Press.
[49] Neumann, J. von and Morgenstern, 0. 1953. Theory of games and economic behavior, Princeton: Princeton University Press.
[50] Nietzsche, F. 1968. Twilight of the idols. Translated by R. Hollingdale. London: Penguin Books.
[51] Nietzsche, F. 1973. Beyond good and evil. Translated by R. Hollingdale. London: Penguin Books.
[52] Nietzsche, F. 1968. The will to power. Translated by W. Kaufraann. New York: Random Books.
[53] Nozick, R. 1981. Philosophical investigations. Cambridge, Massachusetts: The Belknap Press of Harvard University Press.
[54] Nussbaum, M. 1984. Plato on commensurability and desire. Proceedings of the aristotelian society supplementary volume 58, pp. 58-80.
[55] Nussbaum, M. 1986. The fragility of goodness: Luck and ethics in Greek tragedyand philosophy. Cambridge: Cambridge University Press.
[56] Parfit, D. 1984, Reasons and persons, Oxford: The Clarendon Press.
[57] Plato 1956. The Protagoras, In Protagoras and Meno. Translated by W.K.C. Guthrie. London: Penguin Books.
[58] Poliak, R. 1978, Endogenous tastes in demand and welfare analysis. Proceedingsof the american economic association 68, pp. 391-401.
[59] Porter, M. 1985. Competitive advantage: Creating and sustaining superior performance. New York: The Free Press.
[60] Raiffa, H. 1967. Decision analysis: Introductory lectures on choice and uncertainty. Cambridge, Massachusetts: Harvard University Press.
[61] Ramsey, F. 1964. Truth and probability. In K.E. Kyburg (ed.). London.
[62] Rawls, J. 1971. A theory of justice. Cambridge, Massachusetts: The Belknap Press of Harvard University Press.
[63] Rawls, J. 1982. Social unity and primary goods. In A. Sen and B. Williams (eds.) Utilitarianism and beyond. Cambridge: Maison des Sciences de 1’Homme and Cambridge University Press.
[64] Ryle, G. 1984. The concept of mind. Chicago: The University of Chicago Press.
[65] Samuelson, P. 1966. A note on the pure theory of consumer’s behavior. Collected scientific papers of Paul A. Saumelson, vol. 1. Edited by J. Stiglitz. Cambridge: MIT Press.
[66] Samuelson, P. 1966. Consumption theory in terms of revealed preference. Collected scientific papers of Paul A. Saumelson, vol. 1 Edited by J. Stiglitz. Cambridge: MIT Press.
[67] Sandel, M. 1982. Liberalism & the limits of justice. Cambridge Cambridge University Press.
[68] Savage, L.J. 1972. The foundations of statistics. New York: Dover.
[69] Schacht, R. 1983. Nietzsche. London: Routledge & Kegan Paul.
[70] Schelling, T. 1960. The strategy of conflict. Cambridge, Massachusetts: Harvard University Press.
[71] Schick, F. 1984. Having reasons: An essay on rationality and sociality. Princeton: Princeton University Press.
[72] Schon, D, 1983. The reflective practioner: How professionals think in action. New York: Basic Books.
[73] Schon, D. 1984. The crisis of professional education and the pursuit of an epistemology of practice. Harvard Business School 75th anniversary colloquium series.
[74] Schutz, A. 1967. The phenomonology of the social world. Translated by G. Walsh and F. Lehnert. Evanston, Illinois: Northwestern University Press.
[75] Scitovsky, T. 1976. The joyless economy: An Inquiry into human satisfaction andconsumer dissatisfaction. New York: Oxford.
[76] Sen, A. 1973. Behaviour and the concept of preference. In J. Elster (ed.) Rationalchoice, pp. 60-81. New York: New York University Press.
[77] Sen, A. and Williams, B. (eds.) 1982. Utilitarianism and Beyond. Cambridge: Maison des Sciences de I’Hommo and Cambridge University Press.
[78] Shackle, G. 1979 [1952]. Expectations in economics. Westport, Connecticut: Hyperion Press. [79] Sidgwick, H. 1981 [1907]. The methods of ethics. Indianapolis: Hackett Publishing.
[80] Simon, H. 1954. A behavioral theory of rational choice. Quarterly journal ofeconomics 69, pp. 99-118.
[81] Smith, A. 1976. The theory of moral sentiments. Edited and introduction by D.D. Raphael and A.L. MacFie. Oxford: Oxford University Press.
[82] Stigler, G. 1950. The development of utility theory, parts I and II. The journal ofpolitical economy 58, pp. 307-27 and 373-96.
[83] Stigler, G. and Becker G. 1977. De gustibus non est disputandum. American economic review 67, pp. 76-90.
[84] Taylor, C. 1982. The diversity of goods. In A. Sen and B. Williamson (eds.) Utilitarianism and beyond, pp. 129-44. Cambridge: Maison des Sciences de I’Homme and Cambridge University Press.
[85] Taylor, C. 1985. What is human agency? In C. Taylor Human agency and language, pp. 15-44. Cambridge: Cambridge University Press.
[86] Taylor, C.. 1985. The concept of a person. In C. Taylor Human agency and language, pp. 97-114. Cambridge: Cambridge University Press.
[87] Tversky, A. and Kahneman, D. 1981, The Framing of decisions and the psychology of choice. Science 211, pp. 453-8.
[88] Varian, H. 1978. Microeconomic analysis. Now York: Norton.
[89] Varone, C. 1988, The role of metapreferences in the theory of consumer behavior: Is one ranking enough? B.A. thesis, Cambridge University.
[90] Vickrey, W. 1964. Metastatics and macroeconomics. New York: Harcourt, Brace and World.
[91] Vonnegut, K. 1974. God bless you, Mr. Rosewater. New York: Dell.
[92] Wack, P. 1985. Scenarios: Uncharted waters ahead. Harvard Business Review 63:5, pp. 72-89.
[93] Wack, P. 1985. Scenarios: Shooting the rapids. Harvard Business Review 63:6, pp. 139-50.
[94] Walzer, M. 1983. Spheres of justice: A defense of pluralism and equality. New York: Basic Books.
[95] Warnock, M. 1973. Freedom in the early philosophy of J.-P. Sartre. In T. Honderich (ed.) Essays on freedom of action, pp. 1-14. London: Routledge & Kegan Paul.
[96] Weber, M, Economy & society: An outline of interpretive sociology. Edited by G. Roth and C. Wittich. Berkeley: University of California Press.
[97] Weiszacker, C.C. von 1971. Notes on endogenous change of tastes. Journal of economic theory 3, pp. 342-75.
[98] Winter, S. 1964. Economic “natural selection” and the theory of the firm. Yale economic essays 4, pp. 225-72.
[99] Winter, S. 1975. Optimization and evolution. In R. Day and T. Groves (eds.) Adaptive economic models, pp. 73-118. New York: Academic Press.
[100] Wittgenstein, L. 1958. Philosophical Investigations. Translated by G.E.M. Anscombe. Oxford: Basil Blackwell.
[101] Zeckhauser, R. 1986. Behavioral versus rational economics. The journal ofbusiness 59:4, pp. S434-49.
[102] Zeckhauser, R. and Samuelson, W. 1986. Status quo bias in individual decision-making. Harvard University mimeo.
