The paradox of friendship-and-love: distinguishing dishonesty from insincerity à la Adam Smith Open Access

This paper attempts to resolve an inconsistency in Adam Smith’s moral theory of fellow-feeling. It is true that Smith is mainly regarded in modern literature as the father of economics, given his foundational treatise published in 1776, The Wealth of Nations (Smith, 1976b; see Young, 2023; Khalil, 2025e). However, he was mainly known during his lifetime as a moral philosopher in light of his earlier work published in 1759, The Theory of Moral Sentiments (TMS) (Smith, 1976a; Hanley, 2017, 2019; Khalil, 1990, 2025a).

The inconsistency, which this paper uncovers, lies in the first two chapters of TMS. In these two chapters, Smith offers contradictory definitions of fellow-feeling. In Chapter 1, fellow-feeling is about reflecting an original emotion; thus, if the original emotion is sad, the fellow-feeling is also sad. In Chapter 2, however, the fellow-feeling between friends concerning an original sad emotion engenders a pleasant emotion, i.e. contrary to the original emotion.

The inconsistency, referred to here as the “paradox of friendship-and-love,” becomes clearer once we examine how the two chapters use the term “sympathy.” In Chapter 1, the term is used both as “sympathy-as-mirroring” and “sympathy-as-judgment.” The former fellow-feeling acts as the basis of the latter, i.e. the judgment whether the pitch of the reaction to an event is an overreaction. The judge, the impartial spectator, must feel that the reaction of the person under focus is of a proper pitch so that the judge can bring him- or herself to approve the feeling of the person under focus, i.e. engaging sympathy-as-judgment. Such approval performs a function, namely, the approbation of the pitch, and hence, must be of the same currency as the original emotion. In Chapter 2, the term is about mutual sympathy, where there is no judgment of pitch. Mutual sympathy acts as the basis of fondness, tenderness and kindness between two (or more) friends and loved ones^[1].

Smith (1976a, p. 14) noted the difference between “sympathy” (i.e. sympathy-as-judgment) and “mutual sympathy,” namely, in his famous account of reading a book or poem together (see Khalil, 2025c). He even reflects upon the paradox of friendship on the following page. However, Smith’s account and reflection are very brief. He never again returns to discuss the nature of friendship in the TMS—with only one exception (Smith, 1976a, pp. 31-32; see Khalil, 2025d).

To be precise, he does mention friendship again in the TMS but in different contexts. To note only three mentions, the first is in the context of analyzing love, where love inextricably invites heartaches, vulnerability, emotional turbulence and obsessions (Smith, 1976a, pp. 32–34, 38–40; see Khalil, 2025d). The second mention is when Smith conflates “friendship” with acquaintance; hence, we are more delighted to attend joyful weddings than funerals (Smith, 1976a, p. 47). The third mention is in the context of analyzing communal solidarity evident in tribes and traditional societies, where such solidarity stifles the spread of market-based transactions (Smith, 1976a, pp. 220–222; see Khalil, 2025d). However, these analyses do not amount to returning to and discussing the paradox of friendship.

It is no wonder that Smith scholars and commentators have largely failed to notice the paradox. One exception is David Hume (as far as I am aware), who noted the paradox in a famous letter dated 28th July, 1759. Hume observes that the agreeable (pleasurable) emotions that arise when friends converse about a miserable event that happened to one of them or when the audience watches a tragedy on the stage, are always positive. He found this fact “a difficult Problem” because the substrate event is sad and, hence, according to Hume’s theory, sharing a sad event should produce nonpositive feelings (in Smith, 1976a, p. 46, n. 2; in Smith, 1977, p. 43). Hume never solved the paradox. He simply mentions it^[2].

The paper’s contributions are twofold:

1. It explains Smith’s rendition of friendship, namely, how it gives rise to what this paper calls “transcendent utility,” which differs from the everyday “substantive utility”; and

2. It explains Smith’s solution to the paradox of friendship, namely, the proposition that substantive utility and transcendent utility are incommensurable genera of satisfaction.

There are many implications of the proposed solution of the paradox of friendship-and-love. This paper discusses two of them. The first implication concerns business ethics, as firms and organizations at large care about their legitimacy. As Neuhäuser and Siebke (2019) argue, organizations must indeed care because they are not passive actors but rather political actors that shape their legal and institutional environment. Corporate legitimacy supposes that corporations are actors in the sense that they are individuals. This supposition raises ontological philosophical questions regarding whether only people can be considered individuals and, hence, corporations cannot be ethical or responsible actors. This paper cannot investigate this philosophical question (see Rendtorff, 2019, 2020; Khalil, 1997a, 1997b). This paper assumes that corporations and organizations per se are individuals and, hence, expected to be responsible actors in the sense that their private goals advance the public good. Therefore, stakeholders of the public good may judge the action of a business firm as they judge the action of a person (see Haase, 2020). Following this paper’s proposed solution of the paradox of friendship-and-love, stakeholders judge the business firm along two different kinds of spheres.

The first sphere includes the judgment regarding proportionality that turns out to be similar to the rational choice question: Is the action of the organization proportional to the cause? This includes judging whether the action deviates from responsibility as a result of impetuous reaction to an incentive or, further, as a result of dishonesty and cheating. Such deviation is judged according to the legal code or according to the unwritten norm.

The second sphere includes the judgment of whether the action expresses friendship-and-love: Is the actor’s action manipulative – or is it truly expressive of friendship-and-love? This question is about sincerity and insincerity: Is the declared friendship-and-love the true motive behind the choice?

The second implication of the proposed anatomy of friendship-and-love is the explanation of why people find some market-mediated transactions repugnant. This paper uses the term “repugnance” in a narrow sense, namely, to denote cultural norms that find market-based exchanges of some goods, such as human kidneys and acts of voting, to be taboos. The origin of such repugnant transactions has long attracted the attention of philosophers (e.g. Sandel, 2012, 2013) but has only recently attracted the attention of a few economists (e.g. Roth, 2007; Ng, 2019).

There are many other implications that this paper does not discuss. For example, this paper does not review the vast philosophical literature on friendship and love, going back to Aristotle’s views. This paper does not discuss how love in the sense of friendship differs from eros, agape, and philia in ancient Greek thought (Soble, 1998). Nor does this paper delve into modern debates on how friendship may foster favoritism, nepotism, corruption and obstruction of the law (see Koltonski, 2016). It also neither discusses how love is related to sexual desire nor whether love differs radically from friendship (see Helm, 2017; Caluori, 2013; Badhwar, 1993). Instead, this paper studies the anatomy of love in a narrow sense, as merely a more intense form of friendship (see also Hoffman, 1997).

Section 2 presents the paradox of friendship. Section 3 renders Smith’s understanding of friendship-and-love, or, in short, “fondness.” Section 4 explicates the first implication regarding practical ethics, namely, the difference between dishonesty and insincerity, acts that elicit two kinds of ethical disapproval. Section 5 expounds the second implication regarding also practical ethics, namely, repugnant choices that violate the proposed utility incommensurability thesis when the decision-maker buys-and-sells what a culture considers goods that express love, such as human organs, sex and casting of votes. Section 6 concludes.

In Chapter 1, at the start of the TMS, Smith (1976a, p. 10) agrees with Hume on how to characterize fellow-feeling, what he also calls “sympathy.” Sympathy is about the spectator’s mirroring or mimicking the observed feeling of the person principally concerned, the person under focus. This paper calls this mimicking function of fellow-feeling “sympathy-as-mirroring.”

Smith (1976a, p. 11) immediately qualifies sympathy-as-mirroring. It is not the only type of fellow-feeling. He states that fellow-feeling, in addition, involves approbation, i.e. the judgment of whether the observed feeling is proportional to the cause. For Smith (1976a, pp. 47–48), such approbation leads the person under focus to exercise self-command, as she wants her reaction to be a proper pitch that allows the judge to sympathize with her feeling, i.e. approve it as suitable to the cause. This paper calls this approbation function of fellow-feeling “sympathy-as-judgment.”

However, in Chapter 2 that immediately follows Chapter 1, Smith characterizes fellow-feeling completely differently. He indicates his intention via the title of the chapter: “Of the Pleasure of mutual Sympathy” (Smith, 1976a, p. 13). Mutual sympathy is the fellow-feeling experienced by a human being when she listens in a friendly and loving way to the joyful or painful event of another person. Let us focus on the painful event case, as it reveals the paradox of friendship-and-love. Let us say that the person under focus has lost a job and is conveying painful feelings to a friend. The mutual sympathy between the friends makes both feel ironically joyful: the person under focus feels joyful for sharing his or her pain with the friend, and the friend feels joyful for the intimacy that substrate emotion has occasioned.

This mutual sympathy of joy in Chapter 2, even when the substrate emotion is sadness, contradicts the definition of sympathy in Chapter 1. While mutual sympathy in Chapter 2 always occasions joy even if the substrate is a sad event, sympathy in the sense of mirroring and in the sense of judgment in Chapter 1 occasions joy only if the substrate is joyful.

Do these two chapters entail that Adam Smith is inconsistent? A more general question: how could friendship always occasion joy – i.e. the friendship-and-love-paradox – even when the substrate event is sadness?

As argued elsewhere (Khalil, 2010), Smith’s account of self-command amounts to stating and solving one major problem that faces the economic theory of contracts and promise making, namely, the commitment problem. The commitment problem, as expressed in the prisoners’ dilemma and public good games, is the ex post incentive to cheat after ex ante making a commitment (a promise) to cooperate. However, we may conceive the act of cheating as suboptimal even in one shot or finite prisoners’ dilemma game, as it is no different from succumbing to temptation in intertemporal choice (Khalil, 2019). That is, cheating others after making a promise is analytically the same as cheating future selves; hence, it is optimal to cooperate in prisoners’ dilemmas, as is the case in safeguarding the interest of future selves. It is useful to rely on Smith’s mechanics of self-command to see how people stave off the temptation to cheat (whether others in prisoners’ dilemmas or future selves). That is, Smith’s self-command can be reinterpreted as (optimal) compliance with promises made to others and future selves.

Thus, Smith’s “sympathy,” if restricted to the mechanics of self-command, i.e. sympathy-as-judgment, is about how people stave off temptations. This corresponds to Smith’s (1976a, pp. 18, 67) concept of “propriety of action” judgment. Such a judgment is about the first sphere mentioned above. Namely, this paper interprets Smith’s concept as the micro-mechanics of rational decision-making (see Khalil, 2010):

This paper uses rational choice theory in its elementary form (e.g. Becker, 1993): A decision-maker is rational insofar as she reacts to changes in the environment in ways that prevent the decline of her utility, well-being, or satisfaction. She makes decisions while considering the changing incentives or constraints of the environment, taking her preferences, resources, information, and beliefs about the environment as given.

Given this brief definition rational chice and self-command, actions that are impetuous are both nonrational (suboptimal) and unethical. These impetuous actions entail that the decision-maker has failed to ascertain the proportionality of one’s choice vis-à-vis the cause or incentive. The attention to the proportionality criterion entails that the decision-maker must consider all immediate and expected costs, given the ex ante information available at the time of the decision.

This framework, where impetuous deviation from what is proportional is nonrational, can also be applied to non-impetuous decisions such as deception, cheating or any act of dishonesty. Dishonesty amounts to reneging on promises for no emergency reasons – simply out of well-calculated expected higher benefits that appear ex post, i.e. after making the promises or contracts.

We can decipher two reasons why dishonesty is suboptimal and, hence, nonrational. The first is well-known: Succumbing to opportunism may increase gains, but only in the short-run. That is, dishonesty invites non-myopic decline of gains. This reason makes it obvious that dishonesty, similar to impetuous choices, amounts to succumbing to temptation. Hence, dishonesty falls under the umbrella of the analysis of temptations or weakness of will – i.e., as nonrational.

As for the second reason, it is needed to explain a hard case: how could dishonesty remain nonrational when dishonesty actually gives origin to non-myopic gains? This question challenges the postulated ethics/rationality compatibility thesis. This paper undertakes this challenge and affirms the thesis below.

It is important to attend to this hard case – and not simply sweep it under the rug. The critics of the ethics/rationality compatibility thesis usually rely on the classic prisoner’s dilemma game structure: the dominant strategy, which is supposed to be the rational strategy, is obviously defection (i.e., cheating). That is, for such critics, ethics and rationality are ultimately incompatible. The ethics/rationality compatibility thesis crumbles in the case of any kind of cooperation that takes the form of the prisoner’s dilemma game structure.

Even if we establish the ethics/rationality compatibility thesis in the prisoner’s dilemma game structure, such thesis becomes quickly complicated in light of social choice. That is, according to social choice theory, decision-makers have to agree on a policy and hence forge a contract, on the basis of some voting mechanism that aggregates their preferences. This raises thorny issues such as the Arrow Impossibility theorem, where such aggregation is impossible under reasonable conditions. However, social choice theory and its complications fall outside the scope of this paper (see Hooker, 2013).

Once the ethics/rationality thesis is assured, Smith’s sympathy, taken strictly in the sense of sympathy-as-judgment or as self-command that informs “propriety of action,” affirms the proportionality criterion, i.e. rational choice. Smith scholars should find the proposed interpretation of Smith’s sympathy à la rational choice, i.e. sympathy-as-judgment, surprising, if not audacious.

To defend this interpretation (see Khalil, 2010, 2015, 2017, 2019, 2025c, 2023a, 2024a, 2024b, 2025b), Smith’s sympathy-as-judgment of proportionality can be summed up in five steps:

1. if Smith’s “sympathy” in the strict sense (sympathy-as-judgment) is about self-command;

2. if self-command is about executing the proper choice;

3. if the proper choice is the proportionality of the exhibited emotion/action with respect to the cause (incentive);

4. if proportionality is the core idea of rational choice; and

5. then, Smith’s “sympathy” (in this strict sense) is about rational choice.

In any event, the fellow-feeling of sympathy in Chapter 1 – either functioning as simple mirroring à la Hume or as the vehicle of judgment à la Smith – makes the judge feel joyful if the observed feeling is joyful; it makes the judge feel pain if the observed feeling is painful.

Sympathy-as-judgment and mutual sympathy, which Smith discusses in Chapters 1 and 2 at the outset of the TMS, are inconsistent. It is true that both kinds of fellow-feeling are about the concordance of emotions. However, as Schwarze and Scott (2019) argue, the concordance of emotions is a broad phenomenon, which cannot amount to a concept. The concordance criterion is insufficient to highlight the difference between the two types of fellow-feeling.

With sympathy-as-judgment functioning as the mechanics of self-command, the sympathizer might have the same pitch of emotions as the person under focus. This concordance of pitch prompts the sympathizer to judge the original emotion to be of appropriate extent. While such judgment pleases the sense of self-vindication of the person under focus, even when she is experiencing pain, the sympathizer does not experience or feel such self-vindication.

In contrast, with mutual sympathy functioning as the input that produces the joy of friendship, the friend who is engaged in mutual sympathy might not approve of the pitch of emotions of the person under focus. Nonetheless, insofar as the friend is basically not annoying or ego-centered, the two can enjoy the concordance of love and affection. This concordance of emotions entails, by definition, the experience of joy even when the substrate emotion or shared event is sad or painful. Such concordance of love and affection is necessarily mutual, i.e., shared by the person under focus and the fellow who feels with his sadness. The friend, in the first place, does not mirror and judge the original emotion when she attends to and listens to the pain of the person under focus. The friend rahter mirrors and mutually enters the original emotion when she attends to and listens to the pain of the person under focus.

If otherwise, i.e. sharing sad events with friends is simply about mirroring (and additionally judging) the substrate emotion, people would choose not to fellow-feel with friends. The fellow-feeling would only occasion pain (unless the sympathizer enjoys judging the pitch of emotions of others). Hence, the phenomenon of friendship would be restricted to sharing blissful events. However, friendship involves sharing (without judging) painful events – which occasions joy.

Chapters 1 and 2 are inconsistent, posing the paradox:

In Chapter 2, Smith subtly offers a solution. Specifically, mutual sympathy is not about approbation. It is about a different set of preferences, preferences related to friendship, love or, in short, “fondness.”

In Chapter 2, Smith defines friendship as “mutual sympathy,” where two people share the same emotion and are shocked when they do not:

In another short chapter titled “Of the social Passions” (Smith, 1976a, pp. 38–40), Smith uses the terms “[g]enerosity, humanity, kindness, compassion, mutual friendship, and esteem” interchangeably to denote friendship:

This paper uses the term “friendship-and-love” as the input that gives rise to what Smith calls “fondness.” It uses the term “love,” which Smith uses often, as a more intense fondness than “friendship.”

Smith celebrated love in unmistakable statements. Smith recounts how the feeling of being loved is so conducive to happiness that it goes beyond any expected advantage. Even the “rudest vulgar of mankind” demands love:

Smith is explicit in how love promotes mental health to put it in modern terms:

Smith continues and notes the pleasure we experience when we encounter a family enjoying love:

Smith continues and stresses how uneasy we feel if we encounter the contrary:

However, Smith is quick to note the danger of excessive love as obsessive attachment that can turn the other into a fetish (Smith, 1976a, pp. 31, 33, 40, 243). He wrote that love is one of those passions that are “extravagantly disproportioned to the value of their objects” (Smith, 1976a, p. 33). He even called the love that runs between the sexes, especially at a young age, to be “ridiculous” because the lovers “have long fixed their thoughts upon one another.” Indeed, the consequent “passion appears to everybody, but the man who feels it, entirely disproportioned to the value of the object; and love, though it is pardoned in a certain age because we know it is natural, is always laughed at, because we cannot enter into it” (Smith, 1976a, p. 31; see Khalil, 2025d).

Nevertheless, Smith celebrates love. He traces love, particularly the love and affection found among members of the family, to “mutual sympathy” – the same sympathy that unites friends (Smith, 1976a, pp. 220–224). For Smith, the affection of family members is simply more intense than friendship because of “habitual” encounters. That is, the habitual encounter of family members “renders [mutual] sympathy more habitual, and thereby more lively, more distinct, and more determinate” (Smith, 1976a, p. 220). Therefore, since both are underpinned by mutual sympathy, any difference between friendship and love is a matter of the extent of the habitual encounters.

Smith scholars who have studied his economic theory have highlighted how his view of the rising commercial society is embedded in institutions based on moral principles (e.g. Hirschman, 2013; Wight, 2002; Otteson, 2002; McCloskey, 2007, 2008; Young, 2023). Even a few business ethics scholars have enlisted Smith’s economics to advance arguments regarding business legitimacy (e.g. Wells, 2013; Aßländer, 2020).

Regarding business legitimacy, this paper mentioned at the outset that Smith’s theory of mutual sympathy, in juxtaposition to everyday sympathy, suggests a difference between two spheres of ethical approval and disapproval. The first is the ethical sphere that includes judgments regarding proportionality (i.e. sympathy-as-judgment or proportionality judgment). The second is the ethical sphere that includes the feeling of joy arising from friendship-and-love (i.e. mutual sympathy).

The proportionality judgment, by definition, entails ethical approval and disapproval: impartial spectators approve of an action if it is proportional to the cause and disapprove of it otherwise. It is less obvious, if not challenging, to see how the sphere of friendship-and-love entails ethical approval and disapproval as well.

This task is challenging because Smith's text contains two other kinds of approval/disapproval that are unrelated to the analysis of the paradox of friendship-and-love. The first unrelated kind, ignored by this paper, is the “merit of action” judgment. It stands in contradistinction to the examination of proportionality (“propriety of action”). When the impartial spectator examines the merit of action, according to Smith (1976a, pp. 67–69), she is examining whether the motive is benevolent as opposed to toxic. An action is meritorious, for Smith, only if the motive is free from toxic preferences such as self-aggrandizement or malevolence (hatred, envy, schadenfreude). In our case of judging friendship-and-love, and whether it is the subject of manipulation, the manipulation is not necessarily the outcome of toxic preferences.

The second unrelated kind, ignored by this paper, concerns excessiveness and misguided love. Smith (1976a, pp. 38–40) greatly warned his readers to be afraid of love, as it easily leads to obsession, vulnerability and abuse by others. As argued elsewhere (Khalil, 2024a, 2025d), the question of obsession, vulnerability and abuse falls under the umbrella of proportionality conceived in a broad sense, i.e. it ultimately falls under “propriety of action”: Is the extent of love justified by the reward of love?

Such proportionality judgment regarding love differs from proportionality judgment regarding pecuniary/substantive preferences. Indeed, Smith (1976a, pp. 31–32) made this distinction – although not as clearly as one would have hoped (see Khalil, 2025d).

In any case, the point is that the judgment of excessive love or obsession differs from the focus here, namely, to step outside the proportionality criterion altogether and highlight the other sphere, sphere of friendship-and-love. Such stepping out should allow us to see two kinds of approval/disapproval pertinent to difference between the sphere of proportionality per se, on the one hand, and the sphere of friendship-and-love, on the other hand.

To identify the approval/disapproval judgments related to the focus here, i.e. what is pertinent to business ethics and practical ethics at large, let us give examples. The examples express two kinds of approval/disapproval that run along the proposed solution of the paradox of friendship-and-love, i.e. the difference between proportionality judgment per se and joy of friendship-and-love:

Example I gives rise to the following set of questions: was the manager’s decision impetuous and, hence, unfair? Did the manager take into consideration that such a firing, even when fair, may lower the morale of other employees and, hence, lead to a reduction in productivity? Did the manager follow the rules in his firing of the employee? These sets of questions fall under sympathy-as-judgment, the proportionality criterion (specifically, with respect to substantive benefit).

Let us say that the manager in Example I acted with anger, i.e. hastily. He failed to take the proper (proportional) decision, i.e. the decision was suboptimal. What about the case of calculated dishonesty mentioned above? The manager may misrepresent the quality of a product to customers or cut corners with respect to safety at the workplace. These acts are dishonest because they violate either the law or an unwritten norm. Although a dishonest or cheating act may raise the profit margin, it can still be suboptimal along the proportionality criterion. It is suboptimal, as mentioned above, given that such an act of dishonesty undermines the non-myopic, long-term profitability of the firm.

However, what about the case when dishonesty actually enhances expected long-term profitability? This is a hard case – as it undermines the ethics/rationality compatibility thesis mentioned above. However, this apparently hard case turns out to be an impossible case in the first place.

As detailed elsewhere (see Khalil, 1997c, 2025f), cheating in the prisoner’s dilemma game context is impossible to start with. According to the game, people ex ante know that it is supposedly rational to cheat. Thus, no rational agent would ever enter the game to start with. No one would believe non-credible promises, enter into contracts that cannot be fully supported by adequate punishment or adequate retaliation or participate in cooperation of any sort that takes the shape of the prisoner’s dilemma game structure.

Of course, in commercial contracts, such as between a grocer and a customer from out-of-town, people make decisions according to beliefs based on non-credible promises. Such beliefs, what constitute trust that informs the society’s social capital, may turn out to be mistaken and, hence, some people end up poorer while others richer (Khalil, 2003).

However, we cannot analyze such dishonesty via the prisoner’s dilemma game structure, and hence, we cannot accord such “dishonesty,” which is smart on the money, as being rational. We have to invent other analytical tools to characterize trust in commercial contracts that may invite dishonesty that pays – and where standard theory of rational choice lacks the toolbox to explain it. Thus, we can ascertain the ethics/rationality compatibility thesis even in the case of dishonesty that is smart on the money.

In contrast, Example II originates the following set of questions: Is the manager’s true motive the declared caring about the friendship and comradery culture of the organization? Is the manager actually motivated instrumentally to make the employees feel friendly toward each other and happy in order, in the next day, to impose new measures to extract more work without compensation? Is the manager sincere with the declaration regarding caring about the employees – or about the environment, the well-being of the community at large and so on? It is important to distinguish the question about proportionality (Example I) as opposed to sincerity (Example II). It is true, in both cases the manager is motivated by self-interest – not by some toxic preferences. Still, there is a difference between the two examples. If the manager errs in judgment regarding the proper reaction to a tardy employee, the unfairness can be rectified. The manager can rehire the employee and assure the rest of the employees.

In contrast, if the manager errs when she manipulates a Christmas party for an ulterior motive, i.e. substantive benefit, it is much more difficult to rectify the injury. For one thing, once workers realize that they have been manipulated, they cannot ask for compensation, as there was no contractual agreement, as in Example I. Furthermore, the injury is not mainly about the freely offered extra work. It is rather the feeling of infidelity as a result of feeling that one was “stupidly naïve.” Such feeling prompts the employees to harshly judge the manipulative firm as insincere.

Such manipulation or insincerity violates one of Immanuel Kant’s categorical imperatives, what he calls the “formula of Humanity”: you should not use other people as means to your end; people must be respected as ends in themselves (Audi, 2016; see also Rendtorff, 2013; Thiel, 2013; Steigleder, 2019).

While Example I and Example II express, respectively, honesty and sincerity, there are diverse other moral principles and virtues pertinent to business ethics (see, e.g.. Heath, 2013). However, as related to the focus of this paper, namely, on the paradox of friendship-and-love, this paper studies only these two principles: honesty occasioned by sympathy-as-judgment and sincerity occasioned by mutual sympathy.

Sincerity indeed ranks high on the lists of moral codes, including the Confucian code (e.g. Cheng, 2013; Romar, 2013). As Khalil and Feltovich (2018) show experimentally, participants in a game prefer to lose money than make deals and transact with others whose apologies are perceived as insincere. There is a burgeoning literature on how people manipulate some apparent moral acts as a license to do immoral acts, what is called “moral licensing” (Blanken et al., 2015).

Once one feels that the firm’s action is insincere, it is difficult to restore the image of the firm as a legitimate entity. To be the victim of manipulation, one feels betrayal when one’s feeling of friendship and comradery has been abused. Such a feeling of abuse resembles hurt feelings arising from infidelity. In the case where the manipulative actor or the firm was held high in one’s scheme of moral virtue, the abuse could give rise to indignity (Khalil, 2024c).

Stated succinctly, while the impartial spectator disapproves of the deception within the friendship context (Example II), this disapproval differs from the judgment of deception in the commercial contract context. The former involves the pain of betrayal of friendship-and-love, i.e. transcendent loss, while the latter involves only substantive loss.

Besides the implication regarding the two kinds of approvals discussed above, the proposed solution of Smith’s paradox of friendship-and-love has another implication. Specifically, the solution of the paradox promises to explain at least one type of taboo: why do people judge some exchanges among consenting adults, i.e. who find the exchange to be beneficial to both, to be still repugnant? For example, two consenting adults may exchange sex, kidney organs, children-for-adoption or votes in elections for fees determined by market forces of supply and demand. Why do people regard the transaction of these goods repugnant? (see Roth, 2007; Becker and Elías, 2007; Elías et al., 2017; Castro and Elías, 2025).

The judgment of an exchange as repugnant is not the same as judging some tastes as repulsive and, hence, one is squeamish about such tastes. Squeamishness or repulsion is a distaste that arises from the consumption of a bad – such as the consumption of a rotten fish – or what is imagined to be bad, such as the consumption of dog meat for some people. In contrast, repugnance is a judgment of the consumption of an item good that is in itself neither a bad nor imagined to be a bad. The consumption of such an item is judged as repugnant only as a result of the context, namely, because it is exchanged along the rules of market prices or, equivalently, along the rules of quid pro quo.

These rules sanctify the pursuit of substantive utility, i.e. self-interest. Thus, when human kidneys or sex are bought and sold, such items are pursued as commodities, satisfying substantive utility when they are supposed to satisfy transcendent utility. In other words, the root of the judgment of a transaction as representative is the conflation of substantive and transcendent utilities. Such conflation implies that the two genera of utility are commensurable when they are not.

Put differently, the conflation of the two genera amounts to confusing the context of one genus of utility with the other as if the two utilities are made of the same currency, i.e. belonging to a unidimensional utility metric.

Standard economic theory can easily explain repulsion, squeamishness or nausea as the emotion associated with “bads.” Standard economic theory, however, faces difficulty in explaining repugnance, the emotion associated with “goods.” The only way out of this difficulty is to embrace some goods, such as human kidneys and children-for-adoption, as what constitutes friendship-and-love. That is, they cannot be commensurable with market exchange occurring among juridically independent decision makers, who are transacting to satisfy substantive utility.

This paper analyzes what it dubs the “paradox of friendship-and-love.” Hume is probably the only reader of Adam Smith (as far as I am aware) to realize the paradox, which he characterizes in his famous letter to Smith as “a difficult Problem.” However, to use the proverbial sweep-it-under-the-rug solution, Hume walks away from it, dismissing it as an exceptional, non-ordinary case.

This paper asserts that friendship-and-love is a problem, as it is not an exceptional, non-ordinary case. Smith himself celebrates its importance in the TMS. However, the celebration is too brief and never exploited in the TMS. This should not be surprising. Smith’s project in the TMS is to ground the social order on sympathy-as-judgment, not on friendship-and-love. He indeed states that sympathy-as-judgment, specifically “propriety of action” judgment, is the main focus of Part I of the TMS (Smith, 1976a, pp. 18, 67). Furthermore, he states that sympathy-as-judgment, specifically “merit of action” judgment, is the main focus of Part II (Smith, 1976a).

The “propriety of action” judgment affirms self-command: making sure that decision makers do not overreact to incentives. In contrast, “merit of action” judgment disallows actions motivated by toxic preferences such as self-aggrandizement and malevolence (envy, schadenfreude, spite). As such, the social order is based on rationality rather than friendship-and-love.

This paper shows, for Smith, sympathy-as-judgment simply allows the observer to partake in the same emotion as the decision maker does. If the decision maker is in pain, e.g. losing a job or injured in an accident, the observer also feels the pain. Further, the observer may disapprove of it if the person under focus exaggerates the feeling of pain.

In the case of mutual sympathy that informs friendship, the joy of friendship does not necessarily involve judgment. As such, once the two friends share the painful substrate event, both feel the pain as an amalgamation of selves. Such amalgamation occasions, surprisingly, a joyful feeling – joyful despite the fact that the substrate event is painful.

If we ignore the difference between sympathy (for mirroring or judgment) and mutual sympathy, we face a paradox: how come friendship does not lead to the amplification of the original feeling of sadness – as is the case possible with sympathy-as-mirroring? If friendship is mere mirroring, friendship should not exist.

Smith (1976a, p. 14) solves the puzzle in a simple way: friendship must be “another source of satisfaction” – a source that differs from substantive utility. Friendship-and-love gives rise to what this paper calls “transcendent utility.” The solution amounts to the proposition that the substantive and transcendent utilities are incommensurable.

As for the implications of the proposed theory of friendship, namely, the utility incommensurability thesis, this paper highlights two implications. The first is that there is a difference between dishonesty and insincerity. While the dishonest actor and insincere actor are both motivated by self-interest, i.e. not toxic preferences, there is a difference. The dishonest actor simply double-crosses the trust specified as the underpinning institution of the commercial contract, as when the grocer misleads the buyer about the quality of the good. The insincere actor “uses” the other human as means to his end, where the misuse is the manipulation of the friendship of the other to extract resources from her.

The second implication is that the difference between transactions based on commercial contract and transactions based on love should not be mixed or reduced to the same metric as if they are commensurable benefits. It is exactly the mixing of the two benefits that explains what people feel repugnant when an offspring, in one culture, demands compensation for visiting his ailing father, payment for sex or price for his kidney. Societies set up taboos to prohibit such commodification of goods – where the taboos vary with the variation of cultural values across these societies.

If we insist on following the utilitarian line of thinking, i.e. insist that the two genera of utility can be reduced to a single metric, we face one of two outcomes. The first outcome is that we have to endure intellectual pain of being, on the one hand, unable to explain the dishonesty/insincerity distinction and, on the other hand, unable to explain commodification taboos. The second outcome is that we can minimize such intellectual pain by sweeping the paradox of friendship-and-love under the rug à la Hume.

Much earlier drafts benefited from the support of the Konrad Lorenz Institute for Evolution and Cognition Research and the Monash Business School, Monash University. It also benefited from the comments of Kwang Ng, Suhair Ghannam, Jonathan Wight, participants of a seminar at the Doha Institute for Graduate Studies and a conference at the Smith Institute for Political Economy and Philosophy (Chapman University), and two anonymous reviewers. This draft greatly benefited from the help of Lucy Valenta and the assistance of AJE’s Curie (AI) editorial support (https://secure.aje.com/en/curie). The usual caveat applies.

Open Access funding provided by the Doha Institute for Graduate Studies.

Declarations.

Ethical Approval and Informed Consent: Not applicable.

Statement Regarding Research Involving Human Participants and/or Animals: Not applicable.

Funding: Not applicable.

Author’s contribution: Sole author and, hence, not applicable.

Competing interests: The author declares that he has no conflict of interest of any kind.

Availability of data and materials: This paper contains no data and, hence, is not applicable.