8.1 Introduction

The prevailing view of standard economic theory is that economists need not engage in the analysis of moral agency. For purposes of explanation and prediction, we are told, neoclassical economic theory requires that behaviour of the economic agent under investigation should merely conform to a set of axioms whose plausibility is mostly theoretical. To the limited extent that the behaviour of persons corresponds with these assumptions, economic activity can be explained in the framework of rational choice.

In what follows I argue that broader and therefore better understanding of human choice can help improve the quality of economic explanation. While the uniqueness of a human economic agent conceived of as a moral person surpasses the limitations of the text-book economic theory of choice, it nonetheless plays an important role in informing some of its premises.

The view of human agency I want to offer for the consideration of economists is one that is rooted in the Aristotelian-Thomistic tradition, important parts of which, as I will try to show, are shared by Adam Smith. It is a tradition marked by virtue ethics and development of practical wisdom. This tradition’s anthropological claims about human action expose what is missing in the neoclassical account of the economic agent.

The way I approach the moral aspects of human choice is intended to engage critically some core elements of modern economic theory of preferences. The word ‘moral’ in this context refers to human rational agency which encompasses one’s ability to reason, to formulate one’s goals, values and principles of action, and to pursue one’s goals in line with those principles or values.

The structure of the paper is as follows. The next section discusses some problematic elements in the standard economic understanding of human agency and choice. Particular attention is given to the lack of nuance in the treatment of ends and inexistence of a sound theory of ends formation in this theory. The following section encapsulates the essence of what human moral agency meant in the pre-modern virtue-ethics tradition. Directions for possible application of this broader approach to human choice in economics are suggested next. Final section concludes.

8.2 Neoclassical Account of Economic Agency

8.2.1 Economic Agent – ‘Not Human But Important’

From its very inception as a subject of enquiry, economics has been studied as the product of human agency under conditions of scarcity. These were understood to be real actions of real human persons. With its emancipation from the theological and philosophical vision of man, modern economics gradually lost interest in the moral aspects of human action and agency. Already in the seventeenth century, economics and moral philosophy separated into two distinct paths of enquiry. Nonetheless, the human person as a moral agent continued to be treated in one way or another as an important factor of economic analysis. The famous definition of economics proposed by Alfred Marshall in the late nineteenth century retained the spirit of that approach. In his Principles of Economics he called it ‘a study of mankind in the ordinary business of life’ which ‘examines that part of individual and social action which is most closely connected with the attainment and with the use of material requisites of well-being’ (Marshall 1890, 1).

Since around the middle of the twentieth century, however, there has been a tendency in economics (as well as in other social sciences), to avoid explanation of phenomena, which originate in human action, by appealing to the conscious decisions of persons in favour of impersonal factors (Evans 2002, 14). As the subject of economic science and analysis, human beings have increasingly been losing their distinctiveness as persons, as moral agents, being replaced by impersonal economic agents, necessarily non-moral. It is the latter, not the former that is considered the core subject of today’s economic analysis.

In the course of modernisation and specialisation, the discipline of economics emancipated itself from a broader discourse of moral science. As a result many important normative concepts associated with human choice and decision making were left outside of economic analysis. With the arrival of the modern world, modern economics was born whose assumptions of human conduct originated arguably in the ideas about human rationality and reason which were developed in the Enlightenment period (Langholm 1992, 1–10).

The concept of agency employed in standard micro and macroeconomics is one of the rational agent. Rational agency in the economic sense originates directly from the axioms of rational choice theory and it should not be confused with other concepts of rationality employed especially in philosophy or ethics. The neoclassical concept of the economic agent derives from the work of Samuelson (1947). A Samuelsonian economic agent, as Ross (2005, 245) defines him, is simply ‘any system that observes certain consistency conditions in its behaviour, such that it can be interpreted as if it is maximizing the value of a function that maps a system of preferences over commodity bundles onto the real numbers’.

Standard rationality theory, or expected utility theory, is axiomatic in nature, in that it is formulated around a specific set of logical assumptions (completeness, transitivity, independence, and continuity) chosen in order to produce well-defined preference orderings (von Neumann and Morgenstern 1944). The ‘von Neumann-Morgenstern utility function theorem’ states that ‘if these four axioms are satisfied, any set of well-ordered preferences can be represented by a distinct (monotonic) individual utility function’. The utility function is the theory’s representation of the agent. The individual is defined as no more than a collection of preferences (Davis 2006, 70).

Ross argues that nothing in economic theory requires that economic agency be identified with individual persons, because economists use it to build abstract models of firms, nations, labour unions, consortia in auctions, lineages in evolutionary games, and other feedback-sensitive, incentive-driven systems that have no evident psychological properties at all. Economic agents are thus ‘representative’ optimizers, whose ontological status is indeterminate. Davis (2006, 70), for example, notes that there is nothing in the current understanding of choice that requires it to apply to any type of agent in particular:

The representation of choice now dominant in economics, which involves nothing more than maximisation of a formal objective function, applies equally well to any type of agent, whether it be a single individual, a collection of individuals, an animal, a collection of animals, or even a programmed machine.

It is therefore not important for rational choice analysis that the economic agent is a human being, let alone a moral person. The issue of what constitutes the individuality of this agent never arises. It is not a problem that economists find interesting for their purposes. The economic agent is an abstraction, a useful but merely theoretical construct of the smallest unit of activity under analysis.Footnote 1

Economists of the formal neoclassical school grant that they abstract from the over-determined conditions of social reality, by disregarding all but the economic motives of actors in constructing theoretical models. They do not consider it necessary that their models should correspond with concrete human natures or even a general extra-economic idea of ‘human nature’. In short, neoclassical economics, with its focus on abstract rational choice theory, no longer needs a distinctive notion of a person to explain economic phenomena.

8.2.2 The Peculiar Concept of Preferences

Although nothing in the theory requires that the agents in question be persons conceived of as moral agents, this fact alone does not mean that economics is uninterested in individual behaviour. Its interest in individual behaviour, however, is usually limited. Most economists study human choice because it is a building block in the analysis of social phenomena, rather than because they are concerned with explaining individual persons qua individual behaviour. In this they rely on the abstract conception referred to as homo economicus which co-defines rationality and individuality, not in connection with human behaviour, but by reference to the logical-mathematical properties of equilibrium analysis. Thus, rational choice theory is not treated as psychological theory. The theory considers only the formal or mathematical properties of the agent’s preferences. What the latter term should be taken to mean is a formal ordering relation, constructed so as to allow equilibrium analysis. Hausman defines preferences as ‘total subjective comparative evaluations’. He illustrates this with the example: “To say that Jill prefers x to y is to say that when Jill has thought about everything she takes to bear on how much she values x and y, Jill ranks x above y” (2012, 34, italics mine). So: a preference is comparative (x is ranked above y); the comparison is in terms of value; the valuation is subjective (‘how much she values …’); and it takes account of the totality of factors that the individual thinks relevant to the comparison (‘everything she takes to bear on …’).

As total or overall evaluations, preferences are already informed by reflection on what there is reason to do. Nature of reasons for action is not inquired into. The content and rationality of preferences themselves is considered irrelevant to the theory. What matters is that preferences are known and stable over time. In order to guarantee some set of rational choices, preferences must satisfy three other conditions: they must be complete, transitive and independent of irrelevant. These conditions guarantee some ‘most preferred’ subset of the available options which define a rational choice. Included in the ‘most preferred’ subset are those options which bring about the highest level of utility – another concept of purely analytical value which has no psychological meaning.

Preferences are outcomes of comparative assessments rather than inputs into deliberation. The existence of an all-things-considered preference ordering presupposes that the economic agent has been able to compare the available options from different points of view or according to different reasons or motivations that she may have had, in order to choose among those options. Models that explain and predict choices in terms of constraints, beliefs, and total comparative evaluations do not deny that human agency and motivation are complex matters. They instead locate those complexities in theories of preference formation and change and not in theories of choice. Only this way of defining preferences makes it possible to take them as determinants of choice. The separation of deliberation about ends and choice of means, however, is mostly of analytical use and has little to do with how most human decisions are made.

Such separation can only apply to deliberations wherein objects of preferences can be unequivocally compared. And therein lies the rub. In such deliberations the agent has some single, well-defined goal or function (a dominant end); or the differing goals which he pursues have some common factor, such as the satisfaction of a known desire. This approach has no use in the—not uncommon—situations in which one needs to combine two or more different goals, virtues or standards, which he feels he cannot ignore or downplay but which seem to demand incompatible things of him. In the standard economic approach, a person experiencing different motivations, who may find it impossible to compare all options, and also be unable to make a clear-cut decision between them, would be considered a case in which the agent’s preferences are not fully formed (incomplete). And so he or she is incapable of a rational action. An example of such a case is a choice one has to make between integrity or charity, on the one hand, and some other incompatible goal one might have, say, the pursuit of a suitable retirement, or subsistence, or approval of others (Taylor 1985, 236–7). One can compare the strength of one’s desire to have a cup of coffee now with one’s desire to have a cup of tea now, and the degree of respective enjoyments and satisfactions. But how can either of those desires and their satisfaction be compared with one’s desire to be a fine scholar, a good father, a true friend? Such alternatives are incommensurable, much like vacationing at the beach vs. vacationing in the mountains (Finnis 2011, 424), vs. not vacationing at all and having more money in the bank. Each possibility has some intelligible appeal not found in what makes the other appealing. It is often impossible coherently to combine all the demands that we might consider important and valid. An agent can see one reason as outweighing another only if the two reasons are contained within a single evaluative perspective. While admirable in its generality, however, this theory has almost no descriptive content. More complex incidents of human choice which involve incommensurate options cannot be easily explained in the strict terms of rational choice theory.

Thus, economists in the neoclassical tradition begin their analysis of choice at the point where ends are (or are assumed to be) set. In the Humean spirit they conceive of rationality in an instrumental manner: their task is to specify structures that are appropriate means of achieving those ends, whatever they may be. They do not account for ends which emerge from deliberations about incommensurate desires or evolve from the processes of one’s maturation and learning from experience. The main reason for this is the neoclassical economics’ agnosticism about values. The formation and ranking of human ends is intimately connected to values. Desires can only make sense in relation to one’s idea of the good. Otherwise we cannot explain what we find valuable in them, and deciding between two or more conflicting desires is even more problematic. Choice, empirically speaking, involves not merely instrumental but also practical reason. When we try to decide what to do, or when we explain our own conduct or conduct of others, we invoke reasons for action. Yet, the assumption of stable and invariable fundamental preferences makes the discussion about values entirely redundant for economists (in line with the old maxim: de gustibus non disputandum est) and formation of values ceases to be economists’ problem. In the rational choice framework, all values are converted into a single currency of meaningless utility.

It seems paradoxical that economic choice should not be concerned with the way real people decide, choose, and act. Lacking in its account is any meaningful definition of preferences as ends and of preference formation. This indeed is not a satisfactory achievement for a discipline which claims competence in explaining not merely economic phenomena but non-economic issues as well. Some evolutionary economists go as far as to claim that selection and ordering of ends proves to be the most fundamental economic problem and lament the abandonment of the problem by orthodox economics.Footnote 2

Economists explain their ignorance in the realm of human ends with the claim that this is subject for other social scholars’ authority such as anthropologists, sociologists, and philosophers. In what follows I suggest that better grasp of human ends should not be contracted to other disciplines but revisited in the spirit that was dominant in pre-modern economic thought on which its modern successor was built. I propose that the pre-modern thinkers, who did not have a separate notion of an economic agent, but who analysed all human decisions and actions in broader ethical terms, can offer important insights into what is involved in economic choice qua human choice. They show that the former cannot be completely and cleanly separated from the latter without some meaningful loss. Just how much is missed in the simplified account of economic choice can be illustrated by a short sketch of a much more thorough understanding of reason and choice, one that is based on an old and well-established conception of practical reason.

8.3 Pre-modern Economic Thought: Self-Development and Practical Reason

Contrary to the analytical framework of modern economics, it was the focus on the person that located political economy among the moral sciences in the tradition of Western political philosophy. What are now called the human sciences (economics, sociology, etc.) were traditionally referred to as the moral sciences. The mark of this tradition, from Plato and Aristotle, throughout Christian antiquity, the Middle Ages (especially following Aquinas), and the Scottish Enlightenment of Adam Smith, is its placing of economic thinking within the area of moral actions of human beings. It is this tradition which informed an alternative conception of reason to that which prevailed in the Enlightenment. As Hollis (1977,12; italics mine) notes, ‘there was another voice in the Enlightenment, a voice that argued that the power of reason allows humans to master nature, manipulate society, change culture and, indeed, shape our own selves’. The capacity for shaping our own selves is a distinctly human feature which signifies our capacity for deliberative self-determination and indicates a dynamic and undetermined character of human goals. It was given a special focus in classical philosophy and especially in the virtue ethics tradition as the below cursorily discussed works of its important representatives will show.

Virtue ethics is the oldest and broadest stream of ethical theory in the study of human conduct. Introduced by Plato and Aristotle, it was taken up by the Stoics, used by Cicero, and finally adopted into Christianity in its most lasting form by Thomas Aquinas. In this tradition, right action is defined as that which leads to the ‘well-being’ of the individual. The term ‘eudaimonia’ is a classical Greek word, commonly translated as ‘happiness’, but perhaps better described as ‘well-being’ or ‘human flourishing’ or ‘good life’. The essence of eudaimonia consists in one’s striving toward excellence based on one’s unique potential. At the centre of the eudaimonistic ethics is the view of the person as a being conscious of oneself and one’s own will, concerned not only with the question of what one should do but also of what one should and could desire. In the language of modern economics we would say that the person is concerned with what her ends should be, what she should value. Thus, in the Aristotelian-Thomistic tradition, moral behaviour is not only about fulfilling one’s duty or obligation to others and following rules. It is just as much a theory of the formation of the will, and a development of desires and aspirations (Koslowski 2006). One does not need to accept Aristotle’s theory of the will or even regard the will itself as a meaningful concept (there is a powerful modern trend to treat it as obsolete) to recognize some truth in this observation.

Aristotle, Thomas Aquinas, and Adam Smith are three particularly important representatives of this tradition. What connects the three philosophers is an explicit account of human nature characterized by man’s striving for the good and self-formation (‘man’ was then used, but especially in Aquinas’s case, ‘human being’ is what was meant). Their insights offer a potentially valuable addition to what has been only partially recognized by economic utility theory, decision theory, and other rationality studies. They provide building blocks for a richer concept of the economic agent conceived of as a moral person whose preferences are not fixed and stable but forming and changing. The concept of changing and undetermined preferences highlights the dynamic aspect of human agency and suggests that it is not merely satisfaction of preferences but the very process of their formation which makes human agency different from non-human agents and constitutes part of people’s welfare. Contrary to the instrumental means-ends approach of the neoclassical economic theory of choice, this process calls for the exercise of practical reason by the person in question – a distinct feature of human agency.

8.3.1 Practical Wisdom in Aristotle

The great influence of Aristotle not only on ethics, but also on what later came to be called economic thought,Footnote 3 makes his works a natural starting point of enquiry into the moral nature of an economic agent and economic choice. Aristotle is the first thinker to analyse economic life systematically, so that it is seen as ‘embedded’ in the extended fabric of the community. For Aristotle, economic affairs do not have a free-standing status but properly belong to practical philosophy. Unlike modern economics, he treats of economic phenomena such as production, consumption, exchange, in the context of virtues. Aristotle’s anthropology stresses the importance of man’s capacity and propensity to develop and practise virtue, and accounts for a person’s responsibility for acquiring a good character. His ‘economic analysis’, which is usually identified in Book V of the Nicomachean Ethics and Book I of the Politics, is only a sub-section within an enquiry into other, more essential subject-matters (Finley 1970, 5). The most important criterion of evaluation of these various matters appears to be the opportunity for each agent to realize his own potential and thus to fulfil himself in a flourishing life. When reading Aristotle’s reflections on what we call today economic or market behaviour, one cannot resist the impression that economics for him is mostly about human values, concerned more generally about eudaimonia (Pol, I, 9, 1257b, 40-1258a, 2).

What are today referred to as economic pursuits, though important, are for Aristotle simply one among other means toward human flourishing and excellence. One’s endeavours related to securing some level of material status have a functional, not finalistic, nature and it is a necessary, but not sufficient, instrument for attaining a good life. It is true that a good life cannot do without the possession of material goods. Nonetheless, it is not exhausted by the material component, but depends, rather, on a plurality of human dimensions. Present in Aristotle but missing in modern perception of the economic agent is the motivational complexity of every human choice and thus of economic choice, too.

Aristotle is famous for his strongly teleological perspective according to which the highest human good involves activities that are goal-directed and have purpose. Most importantly, the essential end point (telos) is to achieve the best that is within us. As paraphrased by Johnston (1997, 6): ‘The excellence of the human being is thus going to be associated with growth towards some final realization of his or her true and best nature.’

For Aristotle, an agent is roughly defined as a being who can entertain, and take steps to fulfil, ‘reasonable desires’. For him, the rationality that is the distinguishing mark of agency is neither merely formal nor only instrumental. His is the rationality understood as prudence (phronesis). The ‘phronimos’ in Aristotle is the person who has practical wisdom (as compared to theoretical wisdom), full reasonableness (as opposed to perfect rationality). This includes the ability of discerning what things are good for himself and for mankind at the point of choice, rather than merely theoretically. Such a person is the norm of action, both economic and non-economic (NEII.6:1107a1; VI.11:1143b15: VI.5:1140b8-10).

It is because of complexity and multidimensionality of choice that Aristotle, contrary to the standard economic approach appreciates that deliberation about man’s ends does not benefit from the means-end paradigm (Wiggins 1980). Instead, it requires practical wisdom. For Aristotle, it is the mark of the man of practical wisdom to be able to ‘deliberate well about what is good and expedient for himself, not in some particular respect (health, strength) but about what sorts of things conduce to the good life in general’ (NE, Book VI). Deliberation in Aristotle is not primarily a search for means, but rather a search for the best specification of the good that one wants. A nontechnical deliberation is one in which man has a vague description of something he wants—a good life, a satisfying career, a fulfilling hobby, etc. The problem is not to see what will be causally efficacious in bringing this about, but rather to see what really qualifies as an adequate and practically realizable specification of what would satisfy this want. As long as this specification is not available, there is no room for means. This observation is not given enough consideration in the mainstream economics account of human choice.

Aristotle’s account of human choice and deliberation indicates the unfinished or indeterminate character of our ideals and value structure, which are constitutive both of human freedom and practical rationality itself. The life of virtue consists in intellectual and moral virtue, i.e. in the ability to discern what is best and the ability to act on this judgment. In other words, moral virtue serves to bring our actual choices and preferences into line with what our better judgment recommends to us (Veatch 1962, 101). Aristotle’s account of deliberation indicates that the human choice situation cannot always be treated as forming a closed, complete, consistent system.

One of the purposes of wise decision making is to discover what we prefer or what is really of value. For it is only through wise and morally imaginative deliberation that we gain clarity about those objects we have previously deemed to be of value. Wise decision making does not assume that we already have a fixed and absolute set of preferences or values:

Matters concerned with conduct and questions of what is good for us have no fixity. The agents themselves must in each case consider what is appropriate to the occasion (NE 1103b 26).

In Aristotle (and, as will be shown below, also in Aquinas) a person is conceived through his actions as en route toward some completion (Schenck 1976). Aristotle’s good man develops over time (Burnyeat 1980, 69). Virtue is not something we are born with but something we acquire. In a way, the whole Nicomachean Ethics is Aristotle ’s reply to the question of how we grow up to become the fully adult and self-sufficient rational animal that is the end toward which the nature of our species tends. This process of self-completion and all that it entails reveals the human person to be a potential and not a fully actual being.

8.3.2 Aquinas and the Scholastics: The economic Realm as Space for Personal Development

Influenced by Aristotle, the scholars of medieval universities in Europe known as the Scholastics adopted a similar approach to the study of human action. The moral dimension of human behaviour served as a basis of what was later called scholastic economics.Footnote 4 Most medieval writers who ventured into the discussion of economics saw it as a form of behaviour which like all others must be considered within the discussion of human beings as of God’s creation and, as such, morally or law-bound actors. Scholastic anthropology claims that persons are beings capable of moral (self-)rule, whose sense of duty is ‘an important but not the only determinant of their conduct’ (De-Juan and Monsalve 2006; Decock 2013; Langholm 1998).

Although economics was not yet acknowledged as an independent discipline, it formed a consistent body of doctrine according to which economic relations ought to be ruled, by the laws of distributive and commutative justice, always finally by the ‘law of charity’. The economic realm is just one among other spheres of human activity where learning of virtuous conduct takes place. The anthropology underlying the scholastic paradigm in economics is not that of the homo economicus whose goal is the satisfaction of given preferences. Their famous doctrines of usury and just price are ultimately grounded in the vision of a morally responsible man whose will is thought to be autonomous, capable of creating its own laws, or at least ruling itself (De-Juan and Monsalve 2006). That is one way to speak of ‘free will’. But within the hierarchy of laws, the will remains subject to a plethora of heteronomous laws. Those laws include positive human laws, divine law, and natural law (Decock 2013, 523). This vision of human nature implies the possibility to act contrary to right reason and to positive law. Nothing is pre-determined or „given” but has to be reasoned through and decided by each individual.

Among these Scholastics it was particularly Thomas Aquinas who, in Schumpeter’s view, contributed most to establishing the grounds of modern scientific analysis (1954, 8). Summa Theologiae sketches a broad view of Thomas’s understanding of human being in which he further develops Aristotle’s insights concerning moral agency. In Thomas’s vision, as in Aristotle, human action is teleological or purposive, in that man’s goal is to achieve the virtuous life which realizes, so far as possible, one’s potential as a human being. The intellectual and moral virtues perfect the human intellect and appetite in proportion to human nature (ST Ia IIae., q. 62, art. 2).Footnote 5 In Aquinas’s formulation, the human person is naturally ordered toward seeking the true and achieving the good through the operation of the intellect (reason) and the will, even as sin has corrupted this nature. Whether man seeks good or ill, there is a general orientation toward perceived goods. He does not choose what he knows to be fully bad. Whatever man desires, he desires it under the aspect of good (ST I-II, q.1).

Being human, however, means that we can be very much mistaken in our judgements and decisions. Thus, Aquinas’s approach, similarly to Aristotle’s, focuses on man’s imperfect knowledge of the self and his fundamental uncertainty with reference to what he wants and what he should value. For Aquinas, the telos of persons is grounded in self-consciousness and self-mastery. Man always strives for increasingly more complete self-fulfilment: ‘all desire the fulfilment of their perfection, and it is precisely this fulfilment in which the last end consists, as stated above’ (ST I-II, q.1, art. 5). In order to become a good person, one needs to acquire the virtue of prudence, which is covered in Aquinas’s notion of ‘prudentia’: ST II-II q.47 aa 1–7. Prudence enables us to reason well towards the choice of commitments, projects, and actions, and to apply general practical principles to concrete circumstances in order to choose rightly. Prudence involves both the choice of the end sought and the means to attain that end. It calls for reason rather than impulse; and takes counsel from others in the selection process of ends and then of means to achieve the chosen end (Elmendorf 1892, 4). In the Summa, prudentia is described as superior to theoretical knowledge, for prudence perfects the cognitive faculty.

Aquinas’s conception of the person characterized by prudentia resembles Aristotle’s account of the virtuous person. For Aristotle, ‘the virtuous person performs the right action in the right way at the right time on the right objects’ (Rorty 1980, 380), and for Aquinas it is someone ‘really knowing what one is doing, being aware of the circumstances and consequences of one’s actions, with the right conception of the sort of action one is performing’ (Aquinas , question 61, articles 3 and 4). In both accounts the source of action is the human himself. Life of virtue is the gold standard, the end toward which man is striving.

8.3.3 Moral Maturation in Adam Smith

Adam Smith may not be the most obvious thinker to be associated with the Aristotelian-Thomistic tradition. Yet, his thought is very much in line with the concept of the human agent as a moral person developed by the two great philosophers. The sources of his moral enquiry, especially, go back to the ancient tradition of the Greeks and the Romans (Szulczewski 2015, 91). Smith’s essential premise, which he shared with the ancient thinkers, was that moral science could only proceed from a thorough understanding of human nature and human agency was the root of his economic understanding (Oakley 1994, ix). His understanding of man was much broader than the one that is often mistakenly ascribed to him today. Smith believed that many behavioural regularities can be explained better by an understanding of people’s attitude toward actions, rather than their valuation of final outcomes. He notes that many of our choices are ‘not so much founded upon [their] utility,’ but reflect primarily ‘the great, the noble, and the exalted property’ of the action or activity itself. Related to this, he thought that we should never investigate human nature as such, in a void, namely merely theoretically. Even if such an investigation were possible, it would prove disappointing, for it could teach us little about the origins of men’s attitudes. He, in a likewise manner to Aristotle and Aquinas, puts the practical above the theoretical in most things.

Smith did not accept the static view of man as coming into the world already fully equipped to make correct moral decisions. To him, such a conception of man neglects the everyday process by which moral decisions are actually arrived at. His Theory of Moral Sentiments (TMS, henceforth), in particular, shows man’s devotion to self-improvement. It accounts for the crucial process to which all people are subject, that ‘of moral and psychological development from the early stages of childhood to that of the mature moral agent’ (Brown 1994: 95; Montes 2008: 45). Indeed, Adam Smith’s TMS is devoted to the question of how people form moral sentiments. The Smithian individual agent forms his moral sentiments through an informal learning process that is realized in social interactions (Thoron 2016).

Smith posits in TMS and again in WN a common human ‘desire of bettering our condition’ (WN II.iii.28; TMS I.iii.2.i); and implies, though he nowhere states it explicitly, a broad notion of ‘betterment’, including ‘a moral kind that involves improving our character and has nothing to do with acquiring material goods’ (Fleischacker 2004, 63; see also Raphael and Macfie 1984, 9; Heyne 2008, 59–63; Griswold 1999, 130–136; and Otteson 2002, 196–197). The desire for self-improvement, both material and moral, which Smith takes to be natural for human beings, seems incompatible with the posited behaviour of the creature called rational economic man. He saw virtue in one’s self-improvement, not only in material but also, moral aspect. He says, ‘upon many occasions…

[what] prompts us to the practice of those divine virtues is not the love of our neighbour, it is not the love of mankind (…). It is a stronger love, a more powerful affection, which generally takes place upon such occasions; the love of what is honourable and noble, of the grandeur, and dignity, and superiority of our own characters [italics added].

It is not so much the calculation of potential gains, as the desire to do proper things, to do things that would meet the approval of a disinterested by-stander, the impartial spectator (Choi 1990, 293); God, even. Choi (1990, 294) posits, in a Smithian vein, that it is the awareness of the presence of others like oneself and moral judgments made with this awareness that make a person a human being. Knowledge of what constitutes the proper thing is not always obvious and often requires deliberation. In Smith this deliberation is supported by what he calls the impartial spectator: the hypothetical observer who passes positive or negative judgments upon the actions of those around him, directed by virtuous considerations, whether of the intellectual or some other kind; an imaginary construction that individuals call upon as they evaluate their own sentiments and conduct. We endeavour to examine our own conduct as we imagine any other fair and impartial spectator would examine it. The impartial spectator signifies one’s ability to distance oneself from his particular situation and subjective point of view, and to see oneself from a perspective that he can share with others. It also emphasizes the desire for approbation as a key fact of the human condition.

Smith argues in TMS that it is not the love of praise and attention that motivates moral actions, but the love of praiseworthiness. Though we care about the opinions of others, we care much more about the opinions of the impartial spectator and want to avoid his disapprobation. The ‘man within the breast’ is thus often seen as the key to Smith’s account of the faculty of conscience:

The jurisdiction of the man without, is founded altogether in the desire of actual praise, and in the aversion to actual blame. The jurisdiction of the man within, is founded altogether in the desire of praise-worthiness, and in the aversion to blame-worthiness (TMS, III, i, 31–2).

In Storr’s phrasing, the impartial spectator is ‘the imaginary figure that each of us constructs to offer us moral guidance as we negotiate our lives’ (Storr 2013, 3, emphasis added). The use of the word ‘negotiate’ is of great importance here. Many of our decisions need judgment. Judgment, in turn, is an outcome of a particular type of dialogue we exercise with the impartial spectator as we negotiate various aspects of our lives, or – to put differently – as we try to determine what ends we ought to pursue.

The impartial spectator is both a guide as we experience the world and is affected by our experiences in the world. The refinement of the impartial spectator marks our maturation as agents. That is why, in practical reasoning, it is not assumed that we act from a well-defined base of knowledge but rather that this knowledge is constantly in a process of formation, which in turn depends to a large extent on our experience—for instance, as to whether it is a more or less wise spectator. As an outcome of a dialogic process, such judgment is ‘freely subscribed to and freely acted upon, and cannot be predetermined or rule-bound’.

Smith, like Aristotle, understands moral judgment as the exercise of ‘practical wisdom necessary for right actions in particular situations’ (Hanley 2009, 87). He discusses on a number of occasions how our knowledge of our own best course of action comes from our experience as opposed to solely abstract calculation. Smith’s moral agent is not someone who calculates choice based solely on a set of motivations. His capacity of calculation is imperfect, and his motivations are dynamic and constantly under formation. The inclinations to choose one concrete goal over another and the activity of evaluating others’ and one’s own behaviours are both formative processes for Smith, not initial assumptions. This can be seen in his extensive discussions of the formation of knowledge and the formation of desires, values, and interests.

Economic theory treats the formation process as an aside or an afterthought to the judgment about action, while for Smith it is the judgment itself that forms our objectives—meaning the whole process of judging—as it is also the means we take to realize them.

Whereas economists in the neoclassical tradition analyse choice under a full-information assumption (Levy 1995, 305), Smith’s theory of the impartial spectator indicates that any person’s preferences, values, and purposes are neither given nor stable. They are unknown fully at any one moment, but not ultimately unknowable. The epistemic limitation is a result of man’s limited self-experience. The eventual (partial) knowability is a result of careful and ongoing observation of one’s own moral life as if one were a third-party spectator. A critical mark of Smith’s theory is that man needs to learn to see his interests in their true perspective. Finding out what those interests are, however, is a process that economists are unable to capture if they assume given preferences. As Fleischacker (2004, 61, 63) notes, ‘[w]hen we ask after the ‘nature’ of human beings we are looking for what human beings ‘really’ want, beneath the surface trappings (…) Human nature always includes what people aspire to, for Smith; it is never reduced [as in the economist’s version of utilitarianism] to the desires they merely happen to have.’

The view of the human agent that emerges from the above discussion of Aristotle, Thomas Aquinas, and Adam Smith is a picture of a human being directing himself in relation to his own good by practical reason (phronesis in the Greek, and the virtue of prudentia for Aquinas). When a person acts on the basis of practical wisdom it means his choices are embedded in an ongoing process of self-formation. His decisions shape his judgment of his capabilities and interests, which then inform his future decisions. He does not possess complete and accurate knowledge of himself and others, but he is constantly in a process of (self-)development. His wants and needs are not certain and given determinants of his actions, as they are in the mainstream economic assumptions. It is rather his actions which from his unique experience influence his judgment and are conducive to the formation of his future goals.

8.4 Towards a Better Explanation of Choice

Rational choice theory explains decisions and choice ‘as if’ they were brought about by rational economic agents. While successful in domains which do not need to take account of complexity of human choice (e.g. basic consumption decisions) this approach has significant costs for the study of less obvious decisions of the morally constituted person in the economic realm. One can go so far as to say that its limiting assumptions have deprived the human economic agent of almost everything that could make it a person. It has denied the economic agent the ability to learn about what is good for him and to make value choices. The formal strictures of this theory make it impossible for it to capture an image of an evolving self, indicative of a conscious change in one’s values and preferences. In short, it has neglected his capacity of practical reason. For purposes of analytical rigor and predictive power, it has abstracted this capacity away, as if it were not there. Important aspects of human agency which are not conceivable in the rational choice framework are left outside of the picture. Failing to take account of the nature of human ends it has seriously limited its descriptive and predictive capacity.

And yet, although the above analysed factors of human choice do not meet the strictures of economic rationality, they could nonetheless serve as important objects of economic analysis. For they inform many consumer, investment, and entrepreneurial decisions, as well as choices involved in our work relations and relationships with ourselves and the environment and other people. These situations call for decisions which are not—or at least not exclusively—driven by instrumental means-end reasoning. They may be better characterized by the agent’s openness to unforeseen possibilities and learning. As the above necessarily sketchy accounts of the great pre-modern thinkers show, formation of ones ends – the key category in rational choice theory – is not possible without practical reasoning. This is particularly visible in problems of preference conflict, instability and uncertainty about one’s goals.

There are many examples of conflicts among preferences, and of preference change, that lead to behaviour that is very puzzling from the point of view of received economic theory. They show that richer and less rigid conceptions of agency are needed to account for the fact that a person may have not just a single set but competing sets of preferences; or even contradictory preferences or preference sets. Moreover, meta-preferences which rank the preference sets, and a person’s preferences, may oscillate between alternative states. Decision making based on practical wisdom rejects the notion that all alternatives can be compared on a single measure of utility. Rather, this process entails an exploration of the multiplicity of incommensurable values embedded within the available options.

Unlike the abstract homo economicus, persons have the capacity to form, revise, and change their preferences as they see fit.Footnote 6 They can even have no preference at all! While the mainstream theory assumes complete and transitive preferences which implies that all aspects of a choice setting can be weighed off against each other and all alternatives can be ranked in terms of ‘betterness’, people can sometimes refuse to do this. Real-world persons often do not, cannot, or will not reach judgements about some things.

Likewise, the idea of rational economic choice is an inadequate representation of human choice in conditions of an unknown or not fully specified goal. Whereas rational choice theory paints a picture of a goal-oriented agent, guided by known and stable references, human decision making is not always based on a well-defined goal. As the above discussion shows, the moral person seeks to construct a good life for himself and those who depend on him guided by a vision of the good which he is open to revisit based on new experience. Through the exercise of practical reason, he develops and improves his vision of the good and the good life.

The process of self-formation and development of one’s idea of the good should be regarded as part of human moral agency. Obvious general examples are quitting an addiction or taking up a challenge to become physically fit. Moral human agent is someone who strives for self-betterment through conscious formation and change of preferences—meaning, constant refinement of the preferences, and the reasons for (continued) preferring of something or someone over other possibilities. Without enough experience or at least reflection about what the goal entails, we cannot provide an explicit description of how to arrive at it. We often choose from many imagined futures and can rarely expect with certainty that what we choose will work out. Through open-ended involvement in the various practices of our life we learn to exercise a practical judgment, a judgment that becomes operative in practice but which we cannot always give a fully specified or systematic account of. We nonetheless act on it. It enables us to express what is valuable for us and to form ourselves in accordance with what we find valuable.Footnote 7

What makes persons distinct economic (and moral) agents is thus not to be found merely in the content or character or solidity of preferences, but rather in their instability and process of formation. Unlike nonhuman agents, humans possess a reflective capacity which enables them to form their preferences and make judgments about their preference orderings, as well as change these preferences as a result of a change in their values and purposes. In less formal language we can say that only human choice is preceded by attempts to sort out: (1) what is worth doing (choice of ends), and (2) what sort of person to be or become (as a distinct end). This shows that moral agency is not to be defined only by its capacity for altruistic or other-regarding sentiments but also by one’s striving for self-betterment through a conscious formation and change of preferences in line with chosen ends. As such, the human agent (the person) is more complicated and less certain about his preferences than the traditionally-conceived economic agent. Such agent can better be portrayed not by more elaborate interpretations of what brings utility, but by accounting for the uncertainty and instability of his preferences.

This implies, however, the need to relax the two above-mentioned economic assumptions: (1) that preferences are given and (2) that they are stable. In fact, behavioural economists have also questioned these assumptions. Many aspects of human choice, which were traditionally regarded as insignificant in standard neoclassical theory, have been gaining importance. This is challenging the prevailing view that there is no need for economists to delve into the complex issues determining human choice. It is now more readily admitted that better understanding of how people choose can help develop better ideas about what contributes to their welfare and design better policies. A more complex picture of agency has been inspiring departures from mainstream micro- and macroeconomic models.Footnote 8 These new developments, however, stop short of connecting their conclusions to the moral nature of human agency. Not only does the formation of preferences depend strongly on the many psychological and cognitive factors that do not affect choice according to the rational choice approach (the framing-effect of the problem, the predetermined default rules which prefer certain selection decisions, the way risks are conveyed etc.), but it is also affected by conflicts of interests, values, wishes and desires. Clear and obvious long-term preferences, say, good health, can come into conflict with short-term preferences, such as yielding to temptation and having another cigarette. The value of caring for an old parent may be incompatible with long hours spent at work to maintain one’s family. Choices like this are not made easier by correcting for behavioural biases. They instead require practical reason and, indeed, moral wisdom.

Practically speaking, the strictures of rational choice theory also turn out to be particularly difficult for welfare policy. If policy advice is to be based on the promotion of individual well-being, the advice that comes from assuming that people are well-behaved preference satisfiers cannot apply to a world where individuals fail to reveal well-behaved and well-defined preferences. Since real persons fail to behave as predicted by expected utility theory, because their choices reveal incoherent preferences, their behaviour cannot be meaningfully represented by a utility function. Welfare economists thus lose evidence concerning that which promotes their welfare, in all but the most basic or general preference sets (water, shelter, food, clothes…). In other words, they do not have an obvious way to determine what people’s ‘true’ preferences are, or – to be more precise – what they value (Hargreaves-Heap 2013).

Better understanding of how people choose and what constitutes their welfare requires that human agents are understood and explicitly treated as moral, and not merely economic, agents. For, as I tried to illustrate, moral considerations are inseparable aspects of economic activity. They co-determine the choice behaviour of economic agents and thereby direct the outcome of economic activity. Abstracting them away or forcing them into the strictures of rational choice theory limits the extent to which many contexts of human choice can be meaningfully explained and translated into policy.

The above discussion also shows that moral agency cannot be treated as an extra ingredient in the existing framework of economic explanation. Moral agency as understood in this paper is an element which challenges the very fundaments of the rational choice framework. People do not only care about what happens. They are not mere consequentialists. Driven by various non-instrumental considerations, they also care how and for what reason things happen. From the perspective of a moral agent a choice is rational if the chooser can acceptably answer the question ‘Why did you make that choice? or ‘What is the value of your choice?’ In this perspective, inability to grasp and articulate the importance of various goods – i.e. reasons – for one’s conduct is unreasonable, if not plain irrational.

It appears then that part of what it means to be rational involves also deciding about or defining one’s conception of the good and acting on it. Forming a rational response to one’s environment, rejecting preferences one cannot satisfy, or developing preferences that are in line with one’s principles, depends on one’s idea of the good. It depends on one’s values, which one learns and endorses or rejects over time while making choices. An important aspect of human agency is to be found in one’s pursuits of underdefined goals. Persons define themselves and their good (at least partially) through the exercise of this form of agency.

Even in decisions which to an outside observer may seem as primarily driven by reasons based on means-end calculation, people are often motivated by reasons which fall outside of this framework of instrumental reasoning. Consumption which is traditionally regarded by economists as a textbook case of means-ends thinking could instead be seen as an act of discovering one’s needs, one’s values, and even self-formation. Similarly, many entrepreneurial decisions which involve high levels of risk and creativity are difficult to explain ex ante merely as the entrepreneur’s motive of profit maximisation. If economists want to better understand how people form their preferences, outside of the most trivial contexts, they would benefit from a better understanding of the ‘non-economic springs of economic action’ (Hirschman 2013).

8.5 Conclusion

The moral character of human choice is inconvenient for the orthodox economic perspective. There is no place and no logic within this perspective for incomplete and changeable preferences, which in turn precludes any possibility to correctly understand choice. The neoclassical perspective on preferences remains agnostic about the reality of human values which inform their desires, wants and ultimately their choices.

One can wonder, however, whether economists really do not need criteria for distinguishing human individuals as distinct and independent beings, who do not dissolve into social aggregates and patterns. Is there really no use for economic theory to acknowledge that persons count because they are not reducible to economic agents? That their choices also count as their own precisely because the free will and self-formation cannot be reduced to instrumentally conceived economic agency?

Pre-modern virtue ethics tradition and its account of the place of economic activity in a broader context of human moral life provides some valuable insights into how various elements constitutive of the human choice could be better integrated. While rational-choice account of human agency is of purely formal, instrumental and arguably value-free character, the notion of practical reason in the pre-modern tradition of Aristotle, Aquinas and Adam Smith, takes a distinctively normative question as its starting point. It typically asks, of a set of alternatives for action none of which has yet been performed, what one ought to do, or what it would be best to do. It is thus concerned not with matters of fact and their explanation, but with matters of value, of what it would be desirable to do. In practical reasoning, agents attempt to assess and weigh their reasons for action, the considerations that speak for and against alternative courses of action that are open to them. Part of what it means to be rational involves also deciding about or defining one’s conception of the good and acting on it. Insofar as these ‘hidden’ aspects of choice are important in the explanation of choice and the analysis of welfare, they should not be overlooked in economics.