Skip to main content
Log in

A Defense of Modest Ideal Observer Theory: The Case of Adam Smith’s Impartial Spectator

  • Published:
Ethical Theory and Moral Practice Aims and scope Submit manuscript


I build on Adam Smith’s account of the impartial spectator in The Theory of Moral Sentiments in order to offer a modest ideal observer theory of moral judgment that is adequate in the following sense: the account specifies the hypothetical conditions that guarantee the authoritativeness of an agent’s (or agents’) responses in constituting the standard in question, and, if an actual agent or an actual community of agents are not under those conditions, their responses are not authoritative in setting this standard. However, in the account that I provide, the hypothetical conditions can themselves be constructed from the psychology and interactions of actual human beings. In other words, facts about the morally appropriate and inappropriate are determined from hypothetical conditions that––while agents in a given society might have yet to attain them––can be constructed from those agents’ shared experiences. Thus, the account offers both an attainable standard of moral judgment and a standard that can transcend the biases of the society which gave rise to it. I also defend the account against three challenges: (a) ideal observer theories do not offer the right kind of motivation to act on the verdicts of the ideal observer; (b) ideal observer theories cannot explain why the idealization in question is well-motivated and not objectionably ad hoc; (c) the standard used in ideal observer theories cannot be defended upon further reflection, because we would need a non-arbitrary, second-order standard to govern our reflection on the first-order standard of moral judgment.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others


  1. These observations are based in part on Ben-Moshe (2020b, 431–32) & Vallentyne (1996, 101). For further discussion of this type of non-normative response-dependence, see Lewis (1989) & Johnston (1989). Two points are worth noting. First, there is a normative version of response-dependence theory, according to which it is the warranted responses of agents who are under certain conditions that constitute what is morally (or evaluatively) right. See, in particular, McDowell (1998) & Wiggins (1998). I will not be discussing normative response-dependence theories. Second, while I will be arguing in section 5 that the constitutive role of the idealization in Smith’s account is well-motivated, I will generally not be arguing against the view that the responses of agents who are under certain conditions are evidence of, and do not constitute, what is morally right. See Enoch (2009, 322–23) for a discussion of this distinction in the context of constructivism. Thus, the main argument of this paper is not intended to conclusively rule out the realist possibility, according to which the responses of agents under suitable conditions track stance-independent normative facts. Rather, I am appealing, to some extent, to those naturalist anti-realists who already find the constitutivist alternative appealing. I am grateful to an anonymous referee for urging me to clarify this point.

  2. This position was summed up by Firth (1952, 321), who argued that “X is P [an ethical predicate]” means “Any ideal observer would react to x in such and such a way under such and such conditions.” While I will be using Firth as a foil, since he is most identified with ideal observer theory, I will not be following him in using the notion of an ideal observer primarily in order to analyze the meaning of moral terms. Rather, my ambitions are closer to those of Brandt (1959, ch. 10), who uses the ideal observer—or the “Qualified Attitude Method,” as he calls it—primarily in order to provide a standard of correctness of moral judgments. See Carson (1984, 50) for a discussion of this difference.

  3. See Sayre-McCord (1994, 218) for an articulation of epistemic worries that are commonly raised against ideal observer theories. See also Brandt (1955, 409–10). It is worth noting that Kawall (2006) does try to deal with some of these worries. I am skeptical of some of Kawall’s proposed solutions—and of their upshot for the normative import and motivational efficacy of the ideal observer—but engaging with his suggestions is beyond the scope of this paper.

  4. In referring to “sympathy” throughout this paper, I follow Smith’s (and Hume’s) terminology. We would call the phenomenon “empathy,” a term that is a translation of the German Einfühlung and was coined only in1909.

  5. I use “constructed” here in a second-order sense, that is, in relation to the conditions from which the normative judgments in question are made, not in relation to the normative judgments themselves. In other words, it is not the case that the normative judgments in question are constructed under suitable conditions—these judgments are correct if agents’ responses are made under the suitable conditions—but rather the suitable conditions themselves can be constructed from within a given society. Now, realists have argued that if there is unconstructed normative material at the core of one’s view, we have a form of realism in disguise (Enoch (2009, 332) and Shafer-Landau (2003, 42)). Therefore, even if I am using construction in a second-order sense, one could argue that the standpoint that I claim is constructed could have been there all along, but nobody acted on or thought about it. My reply is similar to the one that I noted in fn. 1: my aim is not to conclusively rule out this realist alternative. I am grateful to an anonymous referee for encouraging me to clarify this point. Of course, there might be a worry that the conditions themselves sneak in certain normative dimensions that would make the entire account viciously circular. I discuss this worry in section 3.

  6. TMS is referenced with the relevant part, section, chapter, and paragraph in the Glasgow Edition (Smith 1976).

  7. As I argue elsewhere, the impartial spectator framework is not supposed to be one in which the spectator is a utility-maximizing device (see, for example, TMS III.3.6). In particular, it does not organize “the desires of all persons into one coherent system of desire,” as Rawls (1999, 24) put it, and does not fuse different persons into one person. Rather, when the standpoint of the impartial spectator makes us “see what relates to ourselves in its proper shape and dimensions,” it allows us to understand our interests in the context of our own perspectives and compare them to the interests of others in the context of their perspectives, thus respecting the perspectives of individuals (Ben-Moshe 2021).

  8. Smith believed that our excessive self-love makes it difficult for us to see things from other people’s perspectives (TMS III.4.3).

  9. While Smith argues, in part I of TMS, that when we sympathize with B, we imagine how we would feel in B’s situation (TMS I.i.1.10–13), he argues, in part VII of TMS, that when we sympathize with B, we imagine how B would feel in B’s situation (TMS VII.iii.1.4). As I demonstrate elsewhere, the full development of the latter type of sympathy requires the attainment of the standpoint of the impartial spectator, and so Smith discusses this type of sympathy towards the end of TMS, after he has presented his account of the impartial spectator (Ben-Moshe 2020c). See also Darwall (1998, 268) for an excellent discussion of the importance of the latter type of sympathy in Smith’s moral theory.

  10. This quote is taken from a passage which first appeared in the 2nd edition of TMS, remained with minor variations in editions 3–5, and was replaced by a slightly different passage in the 6th edition (TMS III.2.31–32). The quoted passage can be found in a footnote on pp. 129–130 of the Glasgow Edition (Smith 1976).

  11. I am grateful to an anonymous referee for urging me to clarify this point.

  12. Carson (1984, 51) argues that “if there are only some moral issues concerning which all ideal observers would have the same attitudes, then the ideal observer theory supports an intermediate view between extreme objectivism and extreme metaethical relativism. According to this intermediate view, there is an objectively correct view or judgment concerning some, but not all, moral questions.” I am essentially offering a version of this type of view, albeit instead of referring to agreement or disagreement between different impartial spectators, I am making use of a universal impartial spectator who constraints society- and culture-relative virtues (or different localized impartial spectators).

  13. One might ask about other attributes of the universal impartial spectator. For example, could he be cowardly, cruel, misanthropic, or dishonest? Do we also need to attribute benevolence to this spectator? In reply, recall that this is a notional person and, accordingly, Smith only attributes sufficient knowledge and impartiality, along with sympathy (TMS II.i.2.2 & VII.ii.1.49), to the impartial spectator. The thought that I have developed above is that these attributes would be sufficient to provide the requisite normative constraints on the virtues of a given society. It is worth noting, in regard to sympathy in particular, that Smith argues that just as people are pleased when others sympathize with them and are hurt by the lack of such sympathy, so people are pleased when they are able to sympathize with others and are hurt when they are unable to do so, even when we sympathize with painful sentiments (TMS I.i.2.1–6). See Fleischacker (2012, 301) for a defense of this point. Presumably the impartial spectator would incorporate these aspects of our human nature, and so the standpoint of this spectator would be one from which we are pleased when we are able to sympathize with others. I am grateful to an anonymous referee for pressing me clarify these points.

  14. In this regard, see, for example, Campbell (1971, ch. 6), Griswold (1999, 144), & Sayre-McCord (2010).

  15. One might object that the amount of information the impartial spectator needs in complicated cases—for example, a complicated political decision—might exceed ordinary human capacities. I concede that this might be the case. However, my proposed account would be, in this respect, in the same boat with many moral theories: it seems true of utilitarianism, for example, that the amount of information needed in order to ascertain the right course of action in complicated cases may exceed ordinary human capacities. I am grateful to anonymous referee for raising this worry.

  16. Campbell (1971, 145), for example, argues that the impartial spectator refers “to the normal reaction of a member of a particular group, or of a whole society, when he is in the position of observing the conduct of his fellows.”

  17. See Schliesser (2017, 118–21) for an excellent discussion of counterfactual reasoning in Smith’s account of sympathy.

  18. See, in particular, Campbell (1971, 145) & Raphael (2007, 47–48).

  19. While Smith does write in a footnote that “the present inquiry is not concerning a matter of right […] but concerning a matter of fact” (TMS II.i.5.10), he is clearly looking for a standard that determines what is “fit, and right, and proper to be done” (TMS III.5.5).

  20. One plausible possibility is that those adopting the standpoint of the impartial spectator share their verdicts, and/or their verdicts can be compared to each other by others, in order to formulate universal principles. When such rules have been established, “we frequently appeal to them as to the standards of judgment, in debating concerning the degree of praise or blame that is due to certain actions”; they correct “the misrepresentations of self-love concerning what is fit and proper to be done” and are “commonly cited as the ultimate foundations of what is just and unjust in human conduct” (TMS III.4.11–12). For further discussion of these points, and related issues, see Ben-Moshe (2021).

  21. Even in the case of infanticide, Smith notes that there is a limited context-dependence: “[T]his practice prevails among all savage nations; and in that rudest and lowest state of society it is undoubtedly more pardonable than in any other. The extreme indigence of a savage is often such that […] it is frequently impossible for him to support both himself and his child. We cannot wonder, therefore, that in this case he should abandon it” (TMS V.2.15). For an excellent discussion of the tension between universalism and relativism in Smith’s moral philosophy, see Fleischacker (2011).

  22. These observations about circularity also pertain to the inclusion of sympathy as a characteristic of the impartial spectator: according to sentimentalist reasoning, this inclusion builds on our pre-commitment to associating certain features of human nature (sympathy) with morality in general and with the formation of moral judgments in particular.

  23. Zangwill (2003, 289) mentions the desire to be worthy of approval as a candidate that meets the motivational challenge, but associates it with normative response-dependence theories of value and argues that they are vacuous.

  24. Kawall (2004) argues that Zangwill overlooks the fact that (a) there are different kinds of approval, and (b) the nature of the individuals whose moral approval we seek is important. In connection with (b), I am indeed arguing that approval from the impartial spectator is different from approval from actual spectators; the former, but not the latter, has the requisite normative authority. In connection with (a), instead of focusing on different kinds of approval in Kawall’s sense—he asks us to compare approval towards a moral hero, a beautiful painting, or the skillful tactics of an opponent in a game—I am arguing that the desire to be worthy of approval is different from the desire for approval.

  25. These developmental stages are not articulated by Smith, but are rather my attempt to demonstrate that such a desire can develop over time in individuals. Smith, for his part, thought that the desire to be worthy of approval is “natural,” that is, hard-wired into our psychology: “Man naturally desires […] to be lovely; or to be that thing which is the natural and proper object of love. He naturally dreads […] to be hateful; or to be that thing which is the natural and proper object of hatred” (TMS III.2.1). Moreover, Smith notes that the desire to be worthy of approval is not derived from the desire for approval, and that the two desires are “in many respects, distinct and independent of one another” (TMS III.2.2). However, the desire to be worthy of approval cannot be fully formed at the outset, since the standard of the impartial spectator—the standpoint that determines what is worthy of approval—is not yet in place. The distinction between the two developmental stages, along with the accompanying examples, is borrowed from Ben-Moshe (2020a, 1079 & 1086).

  26. Zangwill’s mention of our folk conception of moral motivation suggests that he believes that the motivational challenge can be met if the motivation in question refers to a normative concept such as duty or rightness or worthiness, rather than, for example, mere approval or self-interest. He does not understand the motivational challenge as being focused on the question of whether the motivation is necessarily successful in attaining what is morally appropriate.

  27. It is worth noting that I have not conclusively shown that most human beings desire to be worthy of approval, even in the modest sense of what I have called “weak approvability.” Providing empirical studies confirming this hypothesis, insofar as they exist, is, alas, beyond the scope of this paper. Moreover, even if the hypothesis is generally true, it is still possible that there are individuals who do not desire to be worthy of approval at all. However, as  I argue elsewhere, apart from the fact that it is questionable whether there are any non-disabled human beings above a certain age who do not desire to be worthy of approval at all, if there are such agents, we may not want to say that they are distinctively human and so we would not know what to make of the application of moral standards to them. Indeed, Smith argues that human beings have developed the desire to be worthy of approval, and not only the desire for mere approval, as the latter “would not alone have rendered [them] fit for the society for which [they were] made”; rather, it is the desire to be worthy of approval that is “necessary” to make human beings “really fit” for society (TMS III.2.7). Therefore, the desire to be worthy of approval is needed in order to make us the type of social beings that we are, and our social systems, or perhaps even our species, are structured so as to instill such a desire (Ben-Moshe 2020a, 1084).

  28. In an earlier paper, I made similar observations about a reconstruction of Hume’s general point of view (Ben-Moshe 2020b, 445-446). It is worth noting that I focus primarily on the idea that our justificatory practices vindicate idealizing, since I believe that this is the most promising strategy for motivating ideal observer theory. However, there may be other alternatives. For example, in a paper that criticizes Enoch’s position, Sobel (2009, 343) argues that the most obvious rationale for idealization is to provide the agent “with a more accurate understanding of what the option she is considering would really be like” (though Sobel’s focus in the paper is on subjectivist (desire-based) accounts of well-being).

  29. Realists like Enoch would argue that our justificatory practices are committed to more than these requirements, namely, to (a) truth and hence also to (b) mind-independent normative facts. My aim is not to take a position about (a) or about whether (a) entails (b). Rather, my aim in the discussion above is to show that the constitutive role of the idealization in Smith’s account is well-motivated, and, furthermore, that two key requirements of our justificatory practices are satisfied. Nevertheless, it is worth noting that while Smith is a meta-ethical sentimentalist—he notes, for example, that “the first perceptions of right and wrong […] cannot be the object of reason, but of immediate sense and feeling” (TMS VII.iii.2.7)—his moral philosophy is not hostile to the idea that moral judgments can express genuine beliefs and hence be true or false. Thus, Smith’s moral philosophy is compatible with the aspirations of Firth’s ideal observer theory, according to which an ideal observer’s reactions determine, as noted earlier, whether an ethical judgment is true or false.

  30. Smith briefly discusses reflective endorsement in connection with Hutcheson’s position, with which Smith disagrees, according to which moral predicates do not apply to the moral sense (for the same reason that sensory predicates do not apply to the senses—sight, for example, cannot be said to be black or white). In this regard, he merely argues that we can use the standpoint of the impartial spectator in order to reflect and pass judgment on the moral sentiments and faculties of other spectators. In fact, we have the ability to override the sympathy we feel towards the actor and assess the moral sentiments and faculties of the spectator judging the actor (TMS VII.iii.3.8–9). And more generally: “Correct moral sentiments […] naturally appear in some degree laudable and morally good” (TMS VII.iii.3.10). Therefore if, upon reflection from the standpoint of the impartial spectator, certain sentiments do not appear laudable and morally good, we have good reason to think that they are incorrect and cannot form the basis for moral judgments. In other words, Smith does not use reflection in order to test and ratify the standard of the impartial spectator itself; rather, he argues that, using the standpoint of the impartial spectator, we can reflect on the moral faculties of spectators and, indeed, on sentiments more generally.

  31. Regarding “ideal,” see TMS III.3.26–29, III.3.38, & III.4.4; regarding “demigod,” see TMS III.2.32, VI.iii.18, & VI.iii.25.

  32. It seems that Sayre-McCord does not think that there is triviality involved here, because, contrary to the impartial spectator approving his own patterns of approval, it is logically possible that he would not approve of us using this standard. However, surely triviality is not merely a function of lack of alternative logical possibilities: if an agent needs a reason to rule out p—as the impartial spectator would need a reason, on pain of arbitrariness, to make the decision regarding his disapproving of us using this standard—but there is no conceivable reason for him to do so, it would seem that p is as trivial as it would be if there were no alternative logical possibilities.

  33. In this regard, see in particular Kornblith (2012).

  34. Smith does argue that there are some situations in which, while we can imagine what the impartial spectator would approve of and closely approximate the impartial spectator’s judgment, we will do so only imperfectly (TMS I.i.5.8). Moreover, as noted earlier, Smith suggests that even the wise and virtuous man can only approximate the impartial spectator’s judgments. Nevertheless, given Smith’s other views about the impartial spectator, it is clear that most of us can become the impartial spectator to the degree needed for this ideal observer to count as our ideal self.

  35. I am following in part the later Williams (1995, 35; 2001, 91) when he argued that p is a reason for A to ϕ only if A could reach the conclusion that he should ϕ by a sound deliberative route from his motivations. For a discussion of this point in connection with Smith’s impartial spectator account—albeit one that does not differentiate between a societal and a universal impartial spectator—see Ben-Moshe (2020a, 1079–80).

  36. I am grateful to an anonymous referee for urging me to clarify these points.


  • Ben-Moshe N (2020a) An Adam Smithian account of moral reasons. Eur J Philos 28(4):1073–1087

    Article  Google Scholar 

  • Ben‐Moshe N (2020b) Hume's General Point of View: A Two-Stage Approach. Pac Philos Q 101(3):431–453

  • Ben-Moshe N (2020c) Making sense of Smith on sympathy and approbation: other-oriented sympathy as a psychological and normative achievement. Br J Hist Philos 28(4):735–755

    Article  Google Scholar 

  • Ben-Moshe N (2021) Comprehensive or political liberalism? The impartial spectator and the justification of political principles. Utilitas. Online First:1–17.

  • Brandt RB (1955) The definition of an “ideal observer” theory in ethics. Philos Phenomenol Res 15(3):407–413

    Article  Google Scholar 

  • Brandt RB (1959) Ethical theory: the problems of normative and critical ethics. Prentice-Hall, Inc., Engelwood Cliffs

    Google Scholar 

  • Campbell TD (1971) Adam Smith’s science of morals. Routledge, Oxon & New York

    Google Scholar 

  • Carson TL (1984) The status of morality. D Reidel Publishing Company, Dordrecht

  • Darwall S (1998) Empathy, sympathy, care. Philos Stud 89(2–3):261–282

    Article  Google Scholar 

  • Enoch D (2005) Why idealize? Ethics 115(4):759–787

    Article  Google Scholar 

  • Enoch D (2009) Can there be a global, interesting, coherent constructivism about practical reason? Philos Explor 12(3):319–339

    Article  Google Scholar 

  • Firth R (1952) Ethical absolutism and the ideal observer. Philos Phenomenol Res 12(3):317–345

    Article  Google Scholar 

  • Fleischacker S (2011) Adam Smith and cultural relativism. Erasmus J Philos Econ 4(2):20–41

  • Fleischacker S (2012) Sympathy in Hume and Smith: a contrast, critique, and reconstruction. In: Fricke C, Follesdal D (eds) Intersubjectivity and objectivity in Adam Smith and Edmund Husserl. Ontos Verlag, Heusenstamm, pp 273–311

  • Fleischacker S (2013) Adam Smith’s moral and political philosophy. In: Zalta EN (ed) Stanford Encyclopedia of Philosophy.

  • Frankfurt HG (1971) Freedom of the will and the concept of a person. J Philos 68(1):5–20

    Article  Google Scholar 

  • Griswold CL (1999) Adam Smith and the virtues of enlightenment. Cambridge University Press, New York

    Google Scholar 

  • Johnston M (1989) Dispositional theories of value. Proc Aristot Soc Suppl Vol 63:139–174

    Google Scholar 

  • Kawall J (2004) Moral response-dependence, ideal observers, and the motive of duty: responding to Zangwill. Erkenntnis 60(3):357–369

    Article  Google Scholar 

  • Kawall J (2006) On the moral epistemology of ideal observer theories. Ethical Theory Moral Pract 9(3):359–374

    Article  Google Scholar 

  • Kornblith H (2012) On reflection. Oxford University Press, Oxford

    Book  Google Scholar 

  • Lewis D (1989) Dispositional theories of value. Proc Aristot Soc Suppl Vol 63:113–137

    Google Scholar 

  • McDowell J (1998) Values and secondary qualities. In: Mind, Value & Reality. Harvard University Press, Cambridge, pp 131–150

    Google Scholar 

  • Newton J (1788) Thoughts upon the African slave trade.

  • Raphael DD (2007) The impartial spectator: Adam Smith’s moral philosophy. Oxford University Press, New York

    Book  Google Scholar 

  • Rawls J (1999) A theory of justice, revised edn. Harvard University Press, Cambridge

  • Sayre-McCord G (1994) On why Hume’s ‘general point of view’ isn’t ideal – and shouldn’t be. Soc Philos Policy 11(1):202–228

  • Sayre-McCord G (2010) Sentiments and spectators: Adam Smith’s theory of moral judgment. In: Brown V, Fleischacker S (eds) The philosophy of Adam Smith: the Adam Smith review, vol 5. Routledge, London & New York, pp 124–144

    Google Scholar 

  • Sayre-McCord G (2013) Hume and Smith on sympathy, approbation and moral judgment. Soc Philos Policy 30(1–2):208–236

    Article  Google Scholar 

  • Schliesser E (2017) Adam Smith: systematic philosopher and public thinker. Oxford University Press, New York

    Book  Google Scholar 

  • Shafer-Landau R (2003) Moral realism: a defence. Oxford University Press, New York

  • Smith A (1976) The Theory of Moral Sentiments. In: Raphael DD, Macfie AL (eds) Liberty Fund, Indianapolis

  • Sobel D (2009) Subjectivism and idealization. Ethics 119(2):336–352

    Article  Google Scholar 

  • Taliaferro C (1988) Relativizing the ideal observer theory. Philos Phenomenol Res 49(1):123–138

    Article  Google Scholar 

  • Vallentyne P (1996) Response-dependence, rigidification and objectivity. Erkenntnis 44(1):101–112

  • Wiggins D (1998) A sensible subjectivism? In: Needs, values, truth: essays in the philosophy of value, 3rd edn. Oxford University Press, New York, pp 185–214

    Google Scholar 

  • Williams B (1981) Internal and external reasons. In: Moral luck: philosophical papers 1973–1980. Cambridge University Press, Cambridge, pp 101–113

    Chapter  Google Scholar 

  • Williams B (1995) Internal reasons and the obscurity of blame. In: Making sense of humanity and other philosophical papers 1982–1993. Cambridge University Press, Cambridge, pp 35–45

    Chapter  Google Scholar 

  • Williams B (2001) Postscript: some further notes on internal and external reasons. In: Millgram E (ed) Varieties of practical reasoning. MIT Press, Cambridge, pp 91–97

    Google Scholar 

  • Zangwill N (2003) Against Moral Response-Dependence. Erkenntnis 59(3):285–290

    Article  Google Scholar 

Download references


I would like to thank audiences at the APA Central Division Meeting (2020) and the Central States Philosophy Association Meeting (2019) for their feedback. I am especially grateful to Derrick Baker, Samuel Fleischacker, Ben Miller, James Rowe, Anthony Rudd, Karsten Stueber, David Sussman, and Alyssa Walker for detailed and penetrating comments. Finally, I would like to thank two anonymous referees for Ethical Theory and Moral Practice, whose excellent comments were of great help in improving the paper.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Nir Ben-Moshe.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Ben-Moshe, N. A Defense of Modest Ideal Observer Theory: The Case of Adam Smith’s Impartial Spectator. Ethic Theory Moral Prac 24, 489–510 (2021).

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: