A missing touch of Adam Smith in Amartya Sen’s Public Reasoning: the
Man Within for the Man Without
Laurie Bréban1, Muriel Gilardone2 and Benoît Walraevens3
Provisional version, november 2014
0. Introduction
This paper aims at introducing a missing touch of Adam Smith in Amartya Sen’s view of “public
reasoning”. Since the publication of “Open and Closed impartiality” (Sen, 2002) and more
extensively in The Idea of justice [IJ] (Sen, 2009), Sen claims to derive his theory of justice from
Adam Smith’s idea of the ‘Impartial Spectator’” in The Theory of Moral Sentiments [TMS] (17591790). According to him, Smith’s concept allows challenging standard approaches of justice, and
more particularly, John Rawls’ theory of social contract (1974), at least on one fundamental
issue: it would involve what Sen calls “open impartiality” as opposed to “closed impartiality”. This
challenging issue that the author pretends to borrow from Smith constitutes the founding
principle of his comparative theory of justice (§1).
Now, Sen’s tribute to Smith’s pioneering concept of impartial spectator gives rise to a set of
criticism. Some of them, related to the history of ideas, concern his interpretation of Smith’s
concept which is considered as unfaithful with regard to the original one (Gilardone, 2010;
Forman-Barzilai, 2010; Bruni, 2011; Pico, 2012; Shapiro, 2011; Ege, Igersheim & Le Chapelain,
2013). Others, more internal to Sen’s analysis, rather concern his use of Smith’s impartial
spectator which would be a weak point of his comparative theory of justice (Shapiro, 2011; Ege,
Igersheim & Le Chapelain, 2013). In the paper, we try to address these two sets of criticism.
We agree with commentators that Sen’s reading of Smith is partial. However, it is worth noting
that his aim, in the IJ, is not to provide a faithful interpretation of Smith but rather to build a
comparative theory of justice “from extending Adam Smith’s idea of the ‘Impartial Spectator’” (IJ:
134). Even if Sen shows concern for Smith’s work, it is only as an inspiration serving his own
project, not as an object of study in itself. And, needless to say that it significantly differs from
1
LED, Université Paris 8 Vincennes-Saint-Denis - 2 Rue de la Liberté, 93200 Saint-Denis - France. E-mail:
[email protected]
2 CREM, Université de Caen Basse-Normandie - 19 rue Claude Bloch - BP 5186 - 14032 Caen CedexFrance. E-mail: [email protected]
3 CREM, Université de Caen Basse-Normandie - 19 rue Claude Bloch - BP 5186 - 14032 Caen CedexFrance. E-mail: [email protected]
1
the one supported by the Scottish philosopher. Sen’s project is normative as it aims at offering a
new theory of justice in order to guide collective choices. On the contrary, in the Theory of Moral
Sentiments, Smith’s approach is rather positive: it aims at explaining the origin of moral
judgments. Obviously, this involves important differences in each author’s conception of the
impartial spectator. While the Smithian concept denotes an abstract observer thanks to which
we judge our own conduct, the Senian concept represents real observers whose point of view
makes our common agreed beliefs evolve toward more justice.
Here, we take for granted these differences and to evaluate Sen’s use of the concept of impartial
spectator with regard to his specific project. This allows us to find a path of reconciliation
between his analysis and Smith’s one. Despite major differences, we show that Sen’s version of
the impartial spectator is not altogether inconsistent with Smith’s analysis. Though it does not
correspond to Smith’s genuine concept or to what the Scottish philosopher sometimes call the
“man within”, it is somewhat reminiscent of another figure from his moral philosophy: the “man
without” (§2).
Now, this bridge between both authors’ analysis does not only render Sen’s project more
conciliable with Smith’s analysis. It also opens the path to address Shapiro’s criticism to Sen’s
project (see Shapiro, 2011). According to us, Smith’s genuine impartial spectator (the “man
within”) could constitute the missing piece in Sen’s analysis of the process which must lead
public reasoning toward more justice. This is the reason why, in conclusion, we propose to
introduce Smith’s “man within” to supplement what we could now reasonably call Sen’s “man
without”.
1. An alternative approach to the impartial spectator for a theory of justice:
Amartya Sen’s tribute to Adam Smith
The aim of this section is to point out the reasons underlying Sen’s project to provide a new
theory of justice for the global age based on Smith’s theory of the impartial spectator. We will
first recall that Sen returns to Adam Smith’s pioneering view of impartiality to support his
project of a comparative theory of justice, conceived as an alternative to John Rawls’ (1971)
transcendental theory of justice (1.1). It is well known that Sen also defends the view that the
Smithian impartial spectator allows an openness that is impossible in Rawls’ (1996) definition of
“public reason” which confines public reasoning to the social contractors. But what is less
understood is that the scope of Sen’s “open impartiality” is larger than a mere involvement of
outsiders in public reasoning. We thus show that Sen’s interpretation of Smith’s concept of
impartial spectator provides an alternative choice procedure based on the comparison of
2
diverse social realizations and lines of reasoning from both far and near, without any negotiation
or search for total agreement (1.2).
1.1. The history of ideas to support Sen’s comparative theory of justice
In order to identify the subject matter of a theory of justice, Sen insists that we need to re-visit
what has been written earlier, and not only take contemporary theories as the only relevant
thinking:
“when you’re dealing with disciplines such as moral philosophy and political
philosophy in particular – with such ideas as the idea of justice –, […] people have
taken different views on it. Plato or Aristotle, or in India Kautilîya, have come up in
different ways at that time and they still remain relevant. And therefore the history
of political philosophy is of considerable interest to contemporary political
philosophy. […] there have been really exceptional thinkers of which Adam Smith is
a major example. John Stuart Mill could be another example; Karl Marx would be
another example” (Baujard, Gilardone & Salles, “A Conversation with Sen”,
forthcoming).
If we look at the history of ideas, Sen (IJ) claims, we can mainly distinguish two traditions of
thoughts on justice inherited from European Enlightenment in the 18th and 19th centuries. The
first one, called “transcendental institutionalism”, dates back to Hobbes, Locke, Rousseau and
Kant, and found contemporary developments and success in political philosophy with Rawls’
theory of “Justice as Fairness”. In this contractarian framework, the emphasis is put on
identifying perfectly just institutions for society (IJ, p. 5). It focuses on perfect justice “rather
than on comparisons of justice and injustice” (ibid, p. 6) and concentrates on “getting the
institutions right”, putting aside the analysis of actual societies (ibid). The second one, dating
back to Smith, Condorcet, Bentham, Wollstonecraft, Marx and Mill, is called the “realizationfocused comparison” approach (ibid, p. 7). Primacy is given here to the investigation and
comparison of social realizations, and to the removal of manifest injustice rather than to the
identification of perfect justice. To the “arrangement-focused view of justice” is opposed the
“realization-focused understanding of justice” (ibid, p. 10). The distance between these
traditions, Sen claims, is “quite momentous” (ibid, p. 7), the transcendental theory addressing “a
different question from that of comparative assessment” (ibid, p. 17)4. Sen’s answer to the
question of knowing what should be the subject matter of a theory of justice is clearly in line
with the comparative approach and its focus on reachable social states. At the very beginning of
4
The relevance of such a dichotomy is growingly debated by Sen’s commentators (Ege, Igersheim& Le
Chapelain 2013 and forthcoming, Kandil 2010, Robeyns 2012). As for us, we will not question Sen’s
belonging to a pure comparative approach, neither the possibility to interpret Smith’s concept of the
impartial spectator within this tradition. On the contrary, we claim that Sen’s refusal of transcendental
elements is the reason for which he appeals to a specific version of Smith’s view of impartiality that we
will not contest, but clarify.
3
the book, he writes that “what is presented here is a theory of justice in a very broad sense”
whose “aim is to clarify how we can proceed to address questions of enhancing justice and
removing injustice, rather than to offer resolutions of questions about the nature of perfect
justice” (IJ, p. ix). Again, in the introduction he states: “in contrast with most modern theories of
justice, which concentrate on the ‘just society’, this book is an attempt to investigate realizationbased comparisons that focus on the advancement or retreat of justice. It is, in this respect, not
in line with the strong and more philosophically celebrated tradition of transcendental
institutionalism”(IJ, p. 8).
The opposition between the transcendental and the comparative approaches soon becomes
under Sen’s pen a confrontation of the dominant, Rawlsian theory of “Justice as Fairness” with
his own that he claims to derive from Smith’s theory of the impartial spectator. It is noteworthy
that Sen does not limit himself to a positive endorsement of the comparative approach. He also
denounces at length what he sees as the limitations of “transcendental institutionalism” in
general, and of Rawls’ theory in particular –which Sen (IJ, p. 9) considers as “the most powerful
and momentous exposition” of this tradition5. Two problems are mentioned (Ibid.): (i) the
“feasibility” of reaching a consensus on a unique, transcendental solution6; (ii) the “redundancy”
of the search for a transcendental solution7. In answering his original question – “what do we
want from a theory of justice?” (Sen, 2006) –, Sen challenges the prevailing view in most of
modern theories of justice (Dworkin 2000, Gauthier 1986, Nozick 1974 in addition to Rawls
1971), according to which we want to identify the demands of the “just” society, or the nature of
“just” institutions. Sen’s endorsement of the comparative approach is no surprise though,
knowing that he made significant contributions to social choice theory (Sen 1970, 1974, 1977,
1995, 1999) and to the capability approach (Sen 1980, 1984, 1985, 1989, 1993), both of them
5
It has to be noticed that Sen’s reading of Rawls must be taken carefully. For instance, Ege, Igersheim& Le
Chapelain (forthcoming) follow Sen to say that the early Rawls belongs to the transcendental tradition, but
criticize Sen’s failure to take adequate note of Rawls’ evolution toward more comparative analysis.
Gilardone (2015) underlines the evolution but also the partiality of Sen’s reading of Rawls during the last
five decades. She also shows that the comparative approach developed by the later Rawls is not taken
seriously by Sen because of its reliance on transcendental features (Gilardone 2015, p. 30). Hawi ?...)
6 Sen (IJ, pp. 12-15) takes an example for illustrating his point. Imagine you have three children who ask
for the property of a flute, using the following claims: Anna thinks she deserves it because she is the only
one to know how to play the flute, Bob argues that he is poor and has no toy of his own, Carla insists on
the fact that she made the flute. How to choose between them? All three make reasonable claims, based on
impartial though different grounds.
7 The best illustration is given by Sen (2012, p. 175) in answering Robeyns’ (2012) criticism: “To
demonstrate the unjust nature of the social divisions between women and men, it is not necessary to
persuade others to agree on the diagnosis of what ‘the perfectly gender just world’ would look like”. He
also adds that “the contrast-with-the-perfectly-just is not a particularly useful way of trying to persuade
the obtuse” (Ibid.). His idea is rather to points to “the prevalence of gross – and alterable – inequalities in
the sharing of household chores, large asymmetries in demands for caring for children, hugely unequal
availability of career advancing-pursuits, big inequalities in decisional powers” (ibid.).
4
being part of that “realization-focused” approach8. However Sen agrees with another important
contemporary view in political philosophy: “the interpretation of justice is linked with public
reasoning” (Sen, 2006, p. 215). And in this matter, Sen (IJ, pp. 44-46) considers that Rawls’s
ideas of public reason and impartiality may be seriously competed with Smith’s view of the
impartial spectator.
Indeed, Sen’s exploration of the history of ideas led him to develop a special interest for Smith
who “had so many insights which are not fully explored yet” (Baujard, Gilardone & Salles M.,
forthcoming, “A Conversation with A. Sen”) and from whom there is a lot more to learn than we
have been able to do:
“[…] my main reason for drawing on TMS for my own little attempt on “the idea of
justice”, is simply that TMS is […] a very central contribution to the theory of justice.
[…] the philosophical insights of TMS have not been explored at all as extensively as
say, the writings of Immanuel Kant like Critique of practical reason or Groundwork
have been explored. Smith remains much less understood on that. And it is, as I
discuss in my book, a matter of extraordinary surprise to me that somebody as
open-minded as John Rawls did not make much use of Smith. […]” (ibid)
It cannot be said that such interest for Smith is anecdotal. We count no less than 132 references
to Smith in IJ and the will to draw on Smith in order to provide an alternative theory of justice is
omnipresent. The most critical excerpt can be found in chapter 2 (“Rawls and Beyond”), just
after a long presentation of the limitations and interests of the Rawlsian conception of justice. In
this excerpt, Sen (IJ, p. 70) clearly states that “[t]he idea of addressing the issue of fairness
through the device of the Smithian impartial spectator allows some possibilities that are not
readily available in the contractarian line of reasoning used by Rawls”. And he particularly
highlights four issues that “the Smithian line of reasoning, involving the impartial spectator”,
may address more easily than the social contract approach (Ibid.):
1) dealing with comparative assessment and not merely identifying a transcendental solution;
2) taking note of social realizations and not only the demands of institutions and rules;
3) allowing incompleteness in social assessment, but still providing guidance in important
problems of social justice, including the urgency of removing manifest cases of injustice; and
4) taking note of voices beyond the membership of the contractarian group, either to take note
of their interests, or to avoid our being trapped in local parochialism.
These four issues are at the core of Sen’s theory of justice and our aim is to understand how
Smith’s contributions prove to be a key in this regard. Unfortunately, Sen does not elaborate
more here on Smith’s insights to offer a better conception of justice. And it has to be
8
On the relationship between his work on social choice theory, capabilities and his approach to justice,
see Gilardone (2015).
5
acknowledged though that the fourth issue mentioned in the excerpt is the one which receives
the more attention, in particular because of Sen’s insistence on a specific feature of what he calls
Smith’s “open impartiality” (Sen 2002, 2006, 2010)9. For instance – and this is very
characteristic –, while Sen (2010, p. 50) announces a presentation of “Smith’s open, realizationfocussed and comparative approach to evaluation”, he merely states that Smith was “concerned
with social realizations (resulting from actual institutions, actual behaviour, and other actual
influences), and did this from a comparative perspective” (Sen 2010, p. 58)10. As usual, Sen
(2010, pp. 59-60) chooses to “separate out one particular feature of the Smithian theory” which
he presents as “quite central” to the theory of justice he advocates: “the domain of points of
views that a theory of justice should try to accommodate. How far should we have to go to get
the impartiality that a theory of justice must demand?”
Since 2002, Sen indeed especially claims to derive from Smith’s impartial spectator a procedure
of “open impartiality”, in which people from outside the focal group can participate in the
debates and can be taken into account in public reasoning on justice. According to us, the
isolation of this particular feature may be explained, but is at the source of a false debate. The
cause is the following: Sen has, for his purpose, disconnected the issue of public reasoning from
the search of the just society, and distinguished two questions: “What is the relevant public?
andOn what questions should the reasoning concentrates?” (Sen, 2006, p. 215). Smith’s concept of
the impartial spectator is largely used by Sen to answer the first question and oppose what he
calls Rawls’s “closed impartiality” – in which only people from the focal, contracting group, are
involved. However, Sen acknowledges that Smith’s influence regarding the other three issues is
not less important. We even claim that they are all four intimately related. For us, the idea of
public reasoning grounded on open impartiality that Sen derives from Smith’s impartial
spectator goes much further than a mere inclusion of outsiders in public reasoning as we will
now see.
9
As a result, commentators also focused on that specific issue, independently of the others (see Clare &
Horn 2010, Forman Barzilai 2010, Fleischacker 2011).
10 It is the same in IJ. We have to read very carefully to find out how Smith helps to address the three first
issues. We found the following: Sen refers in a footnote to the fact that Smith’s contributions to philosophy
and jurisprudence focused on both actual behaviours and institutions (Sen, IJ, p. 69) – the second issue;
and he claims that Smith was “involved in comparisons of societies that already existed or could feasibly
emerge” (Sen, IJ, p. 7) – the first issue – and “the elimination of some outrageously unjust arrangements to
enhance global justice” (Sen, IJ, p. 26) – the third issue.
6
1.2. An alternative choice procedure to reach open impartiality: Smith’s
impartial spectator
Sen’s tribute to Smith in IJ is unsurprisingly related to the notion of impartiality which plays, as
every careful reader knows, a crucial role in Smith’s moral theory. Rawls himself appealed to
Smith’s concept, but Sen (IJ, pp.136-138) is very critical of Rawls’ use of Smith’s impartial
spectator11: “He clearly was not familiar with it, and went on misattributing ideas to Smith
taking him to be mainly a utilitarian, which of course Smith never was” (Baujard, Gilardone,
Salles, forthcoming). In fact there are two issues in Sen’s criticism of Rawls’ interpretation: (i)
the idea of an ideal spectator and (ii) the contract point of view instead of the so-called
utilitarian point of view. These two problematic issues, along with Sen’s alternative proposition
based on his own interpretation of Smith, are the key elements to understand Sen’s third
criticism of Rawls conception of impartiality: (iii) its closeness.
(i) Rawls is blamed by Sen for having completely misunderstood Smith’s concept of impartial
spectatorship firstly because he sees him like an “ideal observer” (Sen, IJ, p. 136). According to
Sen, Smith’s concept relied on actual spectators with their own identity and experience: “Smith
requires the impartial spectator to [...] see what the issues would look like with ‘the eyes of other
people’, from the perspective of ‘real spectators’ – from both far and near (Ibid., underlined by us).
(ii) Secondly, Rawls classifies Smith along with Hume as a classical utilitarian. Sen (IJ, p. 137)
rightly notices that Smith did not argue for sensations of pleasure and pain to be the foundations
of our sense of the good and the right. He also recalls that Smith refused to reduce moral
judgments “to counting pleasure and pain, or more generally, to [...] ‘one species of propriety’”
(Ibid.). So that there is no way that Smith’s impartial spectator be a utilitarian. But Rawls’s
proposal to replace the utilitarian by the social contract point of view does not convince Sen who
prefers coming back to what he sees as being Smith’s original idea: “the invoking of disinterested
judgments of ‘any fair and impartial spectator’” (Sen, IJ, p. 123, underlined by us) which “can
work and enlighten without being either a social contractor, or a utilitarian in camouflage” (Sen,
IJ, p. 138).
(iii) For Sen (IJ, p. 134), Rawls’s social contract point of view also carries the major drawback of
privileging “membership entitlement” in the choice procedure he advocates. It is right that
Rawls also adds that impartiality requires placing those entitled members under a veil of
ignorance in the original position for choosing the basic structure of society and the rules of
11
Although Sen’s interpretation of Smith’s impartial spectator is also controversial, as we will see in the
next section, his criticisms of Rawls’s interpretation are not questioned by commentators (see for instance
Ege, Igersheim& Le Chapelain, 2013, p. 191)
7
justice. However, there is no way that Rawls’s device would prevent from falling in the traps of a
closed – and thus partially blind – impartiality, which would undermine its moral relevance. Sen
(IJ, p. 138-139) highlights three moral problems in Rawls’s idea of the impartial spectator as a
member of a specific social group:
1) “exclusionary neglect”: decisions taken by the focal group can have significant
consequences on people outside that group, who are uninvolved in the procedure.
2) “inclusionary incoherence”: decisions taken by the focal group can have an impact on the
size and population of the group itself.
3) “procedural parochialism”: the veil of ignorance is a safeguard against vested interests,
not against local biases and prejudices.
While Sen’s point is here to provide additional criticisms to the transcendental approach, we
understand that he nevertheless tries to know whether such a procedure based on membership
entitlement could be implemented in the comparative approach framework. Sen’s conclusion is
the following: it is more relevant and useful for a theory of justice at the global age to focus on
“enlightenment relevance” rather than on “membership entitlement”, as Smith did. Indeed, Sen
argues that Smith’s theory of the impartial spectator is a solution for overcoming these three
moral problems. Smith prompts us to adopt the position of “every fair and impartial spectator”,
“from far and near”, and to see things “with the eyes of the rest of mankind” in our deliberations
about justice. The Scottish philosopher’s concept of the impartial spectator, as Sen understands
it, is aimed at questioning and criticizing local, entrenched customs and in this sense that it
favors “open impartiality”. In other words, it is because Sen’s approach is comparative,
realizations-focused and allowing incompleteness that he promotes the wider and open public
reasoning as possible. It is clear that this procedure of choice would be pointless or unworkable
(Clare & Horn 2010) in a transcendental approach. Open impartiality is Sen’s means to answer
to the four requirements of a theory of justice.
Contrary to appearances, the issue here is not so much to answer the question of who should be
involved in the public reasoning, but on what procedure should public reasoning on justice be
based on. Sen finds in Smith’s view of impartiality a means to tear up “the veil of ignorance”, that
characterizes “the original position” from which an ideal observer would tell us what would be a
just society. Sen agrees with the necessity of self-distancing in the formation of judgments for
justice. But to him, self-distancing does not amount to ignorance of the self. It rather involves
comparing one’s view with the ones of other real spectators “from far and near”, drawing on
“Smith’s insistence that we must inter alia view our sentiments ‘from a certain distance from us’”
(Sen 2009: 45). This kind of public reasoning is seen as the best way to reach relevant
agreements towards less injustice of social arrangements, since the broadening of admissible
8
voices may allow “more non-congruent principles to be brought into consideration in answering
a wide variety of justice-related questions” (Sen, IJ, pp. 134-5). On the other hand, we have to
keep in mind Sen’s (1992, p. 9) observation that “the ethical force of a social arrangement that is
backed by a consensus or negotiated settlement of all people involved is clearly absent in the
Smithian model involving the impartial spectator”. While this could be seen as a major
weakness, Sen’s latest contributions show on the contrary that it is the main interest he sees in
Smith’s view of impartiality since he does not consider transcendental solutions as ever
accessible. The idea that we may be wrong in our evaluation of justice and injustice and that
there are no compulsive principles must be acknowledged, not for undermining any collective
decision, but for allowing revision if needed.
To sum up, let us recall what Sen (IJ, p. 134) claims to be the three main advantages of a theory
of justice that would extend Smith’s concept of the impartial spectator: 1) accepting the
legitimacy and importance of the “enlightenment relevance” (and not just “membership
entitlement”) of views from others; 2) a comparative line of reasoning, going beyond the search
for a perfectly just society; and 3) an involvement with social realizations, going beyond the
search only for just institutions. As a result, the opposition between Smith’s open impartiality
and Rawls’s closed impartiality is in fact an opposition between a comparative, real-world and
potentially conflicting or incomplete diagnosis of injustice and a transcendental institutionalist
complete search for justice. Smith’s concept of the impartial spectator “who is able to view our
situation from a critical distance” is viewed by Sen (2010: 50) as a contribution “to a fuller
understanding of the requirements of justice, particularly through an understanding of
impartiality as going beyond the interests and concerns of a local contracting group”. In other
words, Sen develops a comparative approach to justice for collective choices and social
evaluations, based on democratic public reasoning, in which “the eyes of other people” plays a
crucial role in favoring impartiality.
2. A Path of Reconciliation between Sen’s and Smith Analysis: the “Man Without”
With regard to what have just been said, a legitimate question may be raised as whether Sen’s
impartial spectator does actually correspond to Smith’s concept? We agree with most
commentators that such is not the case (see, for instance, Forman-Barzilai, 2010; Gilardone,
2010; Fleischacker, 2011; Shapiro, 2011; Ege et al., 2013). However, it is worth noting that Sen’s
aim, in the IJ, is not to provide a faithful interpretation of Smith but rather to build a comparative
theory of justice “from extending Adam Smith’s idea of the ‘Impartial Spectator’” (IJ: 134). Thus,
even if Sen shows concern for Smith’s work, it is only as an inspiration serving his own project,
9
not as an object of study in itself12. And needless to say that this project significantly differs from
the one supported by the Scottish philosopher, more than two centuries ago. From this
perspective, the question to be answer should rather be: can the impartial spectator serving
Smith’s project really serves Sen’s own project?
2.1. An Impartial Spectator Serving Two Different Projects
Answering such a question implies reviewing, in a first time, the role granted to the impartial
spectator in each author’s project.
Commentators usually underline the difference between Sen’s and Smith’s project using the
former own categories especially his distinction between what he calls the “transcendental” and
the “comparative” traditions or between “open” and “closed” impartiality. This generally leads
them to reject the possibility of a Smithian impartial spectator for Sen’s project13. Here, we won’t
enter into the question to know whether there is a kind transcendental dimension or a
conception of open impartiality in Smith’s analysis. We are rather going to emphasize another
grid which allows explaining part of the uneasiness that the reader feel going from the TMS to
the IJ. There is an important difference between Smith’s and Sen’s projects that, to our
knowledge, has never been mentioned in the literature and which concerns the modalities of
their discourse: either positive or normative. This surely plays the part in the difference in what
both call an: impartial spectator.
Let’s start with Smith’s project. In the TMS, the author’s discourse is mainly positive (see
Campbell 1971, Dellemotte, 2011; Evensky, 1987 and Heilbroner, 1982)14. As observed by
Dellemotte (2011: 2237), Smith explicitly presents his moral philosophy as a positive theory of
moral behavior which aims at explaining the origin of our moral judgments. Dellemotte’s
observation relies on two elements: (i) on a footnote of the TMS in which the author makes clear
that the “inquiry” carries out in the book “is not concerning a matter of right […] but concerning
a matter of fact” (TMS, II, i, 5, footnote 10: 77; our italics); (ii) on the complete title of the book
which leaves no doubt about its purpose:
“The theory of moral sentiments. Or, an essay towards an analysis of the principles by
which men naturally judge concerning the conduct and character, first of their
neighbours, and afterwards of themselves.”
12This is true for the history of ideas in general and for his classification of the philosophical traditions
between transcendental and comparative approaches (See Ege, Igersheim, Le Chapelain, forthcoming).
13 This is typically the approach of Ege et al. (2013) according to whom there would be a transcendental
dimension in Smith’s analysis, that Sen would not have been able to grasp, and that would definitively
separate his project from Smith’s.
14 Recent scholarship has nonetheless emphasized the normative side of Smith’s TMS. See especially
Hanley (2009).
10
It is well-known that this “analysis” relies on Smith’s famous principle of sympathy15. However,
in order to explain how “men naturally judge concerning the conduct of themselves”, the author
has to call for an additional concept:
“We either approve or disapprove of the conduct of another man according as we
feel that, when we bring his case home to ourselves, we either can or cannot entirely
sympathize with the sentiments and motives which directed it. And, in the same
manner, we either approve or disapprove of our own conduct, according as we feel
that, when we place ourselves in the situation of another man, and view it, as it
were, with his eyes and from his station, we either can or cannot entirely enter into
and sympathize with the sentiments and motives which influenced it. We can never
survey our own sentiments and motives, we can never form any judgment
concerning them; unless […][w]e endeavour to examine [them] as we imagine any
other fair and impartial spectator would examine it.” (TMS, III, 1: 109-10)
Thus, the “impartial spectator” is the concept that Smith introduces to describe how individuals
achieve to remove themselves from their “natural station” (meaning: “partial station”16) in order
to judge of their own behavior17. He consists in an abstract figure that the author often calls “the
supposed impartial spectator”18 and who represents, as Sen rightly observes, “the eyes of other
people”. Indeed, “[w]hatever judgment we can form concerning [our own sentiments and
motives]”, Smith says, “[this judgment] must always bear some secret reference, either to what
are, or to what, upon a certain condition, would be, or to what, we imagine, ought to be the
judgment of others” (ibid). As a result, the Smithian impartial spectator embodies what we
imagine to be the “judgment of others”. By the way, Smith often refers to him as a “judge”19 or an
“arbiter”20. He constitutes our moral reference and it is shaped thanks to our sympathetic
interactions (see infra, 2.2)21.
15
For a presentation and discussion of Smith’s concept of sympathy, Morrow (1923) remains a relevant
introduction.
16 On the distinction and the interaction between an individual’s “natural point of view” and the “impartial
spectator point of view”, in the TMS, see Bréban, 2014.
17 This process of distancing is well summarized at the end of the same chapter: “When I endeavour to
examine my own conduct, when I endeavour to pass sentence upon it, and either to approve or condemn
it, it is evident that, in all such cases, I divide myself, as it were, into two persons; and that I, the examiner
and judge, represent a different character from that other I, the person whose conduct is examined into
and judged of.” (TMS, III, 1: 113)
18 See, for instance,TMS, III, 2: 130-1; 3: 134; VI, ii, 1: 226; iii, Conclusion: 262.
19See, for instance, TMS, III, 1: 110, 113; 2: 130; 3: 134, 137; VI, ii, 2: 227-228; iii: 245, 247; Conclusion:
262.
20See, for instance, TMS, III, 2: 130; 3: 137; VI, ii, 2: 227; iii: 247; Conclusion: 262.
21 It is noteworthy that the impartial spectator is not people’s only moral reference. Smith provides us
with a hierarchy of moral judgments. The “inferior tribunal” represents the common degree of morality in
society or the usual propriety of conduct. There is also a “higher tribunal”,that of the impartial spectator,
representing “perfect propriety”. However, it does not mean for Smith that the impartial spectator will
provide perfect moral judgments. It is only a “semi-god within the breast”, and its human origin makes it
prone, sometimes, to corruption (from real spectators). The perfection of moral judgment is the privilege
of “a still higher tribunal”, that of God, who is omniscient and perfectly informed.
11
Let’s turn now to Sen’s project. Obviously, contrary to Smith, Sen’s discourse is normative. As we
have seen in the previous section, the IJ prescribes an alternative collective choice procedure,
based on public reasoning, which might lead a society towards less injustice. As for the impartial
spectator, he comes to respond to one of the four requirements of such a procedure: the
requirement, for a particular group of people, of self-distancing while making judgments about
justice (see supra, 1.2). He is the one “who is able to view [the situation of the focal group] from a
critical distance” (Sen 2010: 50). Thus, the impartial spectator is viewed by Sen as the mean for
public reasoning to go beyond the particular interests and concerns of the group.
Up to this point, the Senian impartial spectator may be considered as a possible normative
extension of the Smithian concept. However, this would not take into account two major
differences between them. Indeed, the Senian impartial spectator is (i) neither an abstract figure
(ii) nor a (moral) reference, as it is the case in Smith’s analysis. Let’s deal successively with these
two issues.
(i) With regard to the first issue, the Senian impartial spectator is not an abstract but a real
spectator involved in the public reasoning of a group although “not necessarily (indeed
sometimes ideally not) belonging to the focal group” (Sen 2009: 123). His role can be
endorsed by outsiders (the distant neighbor22) as well as by people from within the
group (the dissident) (see Sen 2009: 123). Whatever be, the impartial spectator must be
able of distancing vis-à-vis the social norms. This leads us to the second issue.
(ii) In no case, Sen’s impartial spectator should be seen himself as a reference regarding
judgments for justice. He does not come as an “arbitrator” (Sen 2009: 131) who would
tell the group what would be the proper decision of justice (note that this would go
against the spirit of the comparative approach defended by Sen). Sen’s impartial
spectator is not impartial because he is able to make “just” judgments (they may be just
or unjust). He is impartial because he has distinct experiences, prejudices and beliefs
from the focal group which render him able of distancing vis-à-vis its social norms.
Thus, he opens the path for questioning standard, established local reasoning and
practices, and then, for the modification of beliefs concerning justice. His different
22The
fact that those we call the distant neighbors are not directly concerned by the decision to be taken
does not mean that they cannot be indirectly affected by it. Indeed, Gilardone & Baujard (2013: 21)
underline that Sen urges us to consider judgments from “outside” as admissible voices, both because of
the enlightenment that their perspectives may provide for those “inside,” and because they might “bear
some of the consequences of decisions taken in that particular polity” (Sen 2009: 134). We can notice a
kind of tension here since the supposed disinterestedness of outsiders might be tone down whenever
their own interests are at stake. This does not entail ruling out their judgments in that case, but
acknowledging the potential conflict of interests that might drive their judgments. Further, it might help
us to see that others are also impacted by our decision. This is another kind of enlightenment.
12
perspective, says Sen, “may help us to achieve a less partial understanding of the ethics
and justice of a problem” (Ibid., our italics). Now, following the author’s understanding
of impartiality, public reasoning should not involve one impartial spectators but as
many as possible and, most importantly, as different as possible. Actually, it should
involve, says Sen, “a wide variety of viewpoints and outlooks based on diverse
experiences from far and near, rather than remaining contented with encounters –
actual or counterfactual – with others living in the same cultural and social milieu, and
with the same kind of experiences, prejudices and beliefs about what is reasonable and
what is not, and even what is feasible and what is not” (Sen 2009: 44).
Thus, unsurprisingly, this review of the role granted to the impartial spectator in each author’s
project leads to conclude that the Senian concept is far removed from the Smithian concept so
that, on first view, it is difficult to see how the latter could serve Sen’s project. However, such a
perspective on each author’s project also leads to foresee some possible bridges between them.
Despite major differences, Sen’s concept of impartial spectator is not altogether irreconcilable
with Smith’s analysis. This is what will be shown in the next subsection.
2.2. The “Man Without”: a Way to Reconcile Smith and Sen
Though he does not correspond to Smith’s concept, Sen’s impartial spectator is somewhat
reminiscent of another figure from the TMS: the “man without” (TMS, III, 2: 130-1). Following
Smith, the man without is the figure which represents the actual spectators of our own conduct
as opposed to the (supposed) impartial spectator that the author also calls the “man within”
(Ibid.).
Indeed, real spectators also play an important role in Smith’s moral philosophy. They take part
in the development of our moral conscience which arises from our social interactions23. Through
social interactions, we find that others (real spectators) judge of our behaviors exactly as we
ourselves, as a spectator, judge of their behaviors. We then become anxious to know, says Smith,
“whether to them we must necessarily appear those agreeable or disagreeable creatures which
they represent us” (TMS, III, 1: 112). That is how we are led to adopt another perspective on our
own behaviors, that is, what we imagine to be the eyes of other people24. In others words, this is
23
On this point, see Raphael and Macfie (1976: 16).
“We begin, upon this account, to examine our own passions and conduct, and to consider how these
must appear to them, by considering how they would appear to us if in their situation. We suppose
ourselves the spectators of our own behaviour, and endeavour to imagine what effect it would, in this
light, produce upon us. This is the only looking-glass by which we can, in some measure, with the eyes of
other people, scrutinize the propriety of our own conduct." (TMS, III, 1: 112).
24
13
from our interactions with real spectators that the “man within” (which enables us to judge of
our own behavior) arises. This leads Smith to conclude that, “[v]irtue is not said to be amiable, or
to be meritorious, because it is the object of its own love, or of its own gratitude; but because it
excites those sentiments in other men.” (TMS, III, 1: 113).
At this stage, an important specification should be made. As stressed by Raphael and Macfie
(1976: 16), for Smith, when we judge our own conduct we do not simply observe the actual
judgment of the “man without”’. We imagine what we should feel if we ourselves were the
spectator of our conduct. We thus appeal “to a much higher tribunal” (TMS, III, 2: 130) 25. There
is indeed a major difference between the “man within” and the “man without” which makes the
former a more reliable judge than the latter: the man within, as our representation of what
should be the others’ judgment, is a well-informed spectator. He possesses information about
our conduct and the motives that influenced it which lack to the “man without”. For this reason,
Smith refers to him as “the supposed impartial and well-informed spectator” (TMS, III, 2: 130) as
opposed the “man without” that he depicts as “ignorant” and potentially “weak” (TMS, III, 2:
131). Indeed, the latter could either be impartial (having no particular connection with us) or
partial (having a particular connection with us or being influenced, in his judgment, by others
criteria than moral ones26).
Now, and this is the second point, despite his imperfections, the man without still keeps an
important role once our moral conscience is developed. Smith claims that there are “some
extraordinary occasions” on which “the testimony of the supposed impartial spectator cannot
always alone support him” (TMS, III, 3: 134). This is the case, for instance, when we are doubtful
about the accuracy of our own judgments about our conduct. In these doubtful cases, the
judgment of the “man without” is of principal consequence, as a proof of the morality (or
immorality) of our conduct (see TMS, III, 2: 122-126).
But most importantly for us, Smith addresses some others, more dramatic, occasions on which
the “man without” or, what he also calls “the real and impartial spectator”27 is “at a great
distance”. Basically, they correspond to situations in which we have no more interactions with
others. Now, “In solitude”, the author says, we are apt to be too partial,
25
“But though man has, in this manner, been rendered the immediate judge of mankind, he has been
rendered so only in the first instance; and an appeal lies from his sentence to a much higher tribunal, to
the tribunal of their own consciences, to that of the supposed impartial and well-informed spectator, to
that of the man within the breast, the great judge and arbiter of their conduct” (TMS, III, 2: 130).
26 These other criteria can be wealth or power. Smith addresses with this specific issue in his chapter: “Of
the corruption of our moral sentiments, which is occasioned by this disposition to admire the rich and the
great, and to despise or neglect persons of poor and mean condition”(TMS, I, iii, 3: 61)
27 See TMS, III, 4: 156.
14
“we are apt to feel too strongly whatever relates to ourselves […] The man within the
breast, the abstract and ideal spectator of our sentiments and conduct, requires
often to be awakened and put in mind of his duty, by the presence of the real
spectator” (TMS, III, 3: 153; our italics)
This observation leads Smith to a recommendation, one of the few that he makes in TMS:
according to him, in adversity as well as in prosperity, we should not stay “in the darkness of
solitude” or in the sole company of people having particular connection with us28. We should go
“to the day-light of the world and of society”. We should face people having no particular
connection with us. We should confront to the point of view of real impartial spectators on our
situation (TMS, III, 3: 154). This recommendation actually expresses the dramatic consequences
that the author grants to these situation in which the man without is at a great distance, e.g.,
situations in which our morality “is never so apt to be corrupted” (TMS, III, 3: 154).
To illustrate the dramatic consequences of such situations, Smith gives an example which, in a
sense, echoes Sen’s concerns in the IJ. In this example, the author transposes his analysis at the
individual level to the level of nations by depicting the conduct of two nations at variance. In
such case, he says, “neutral nations are the only indifferent and impartial spectators [the only
real impartial spectators]”. Unfortunately, “they are placed at so great a distance that they are
almost quite out of sight”. It results that, inside and between the two nations at variance, “the
laws of justice are very seldom observed” (see TMS, III, 3: 154-5).
Though Smith does not make explicit recommendation from this specific example, it is not
unreasonable to imagine that if there was an opportunity for the nations at variance to
communicate with neutral nations, they would be more impartial. Now, the evocation of such an
opportunity is somewhat reminiscent of Sen’s idea of public reasoning which should include
outsiders. As a result, it is not altogether absurd to consider Sen’s impartial spectator as a
normative extension of Smith’s “man without”.
3. Concluding Remarks: The Man Within as a Missing Piece of Sen’s Open
Impartiality?
Going back to the Theory of Moral Sentiments does not only allow building some bridges between
Sen’s and Smith’s analysis. It also opens the path to address Shapiro’s criticism to Sen’s project
(see Shapiro, 2011). We agree with the commentator that the Senian concept of impartial
spectator leaves outstanding issues. One fundamental issue is that it does not explain how
impartial spectators could make common agreed belief of a focal group evolve toward more
28
Smith uses to the words “indulgent and partial spectators” to denote these individuals having particular
connection with us (see TMS, III, 3: 154).
15
justice. But contrary to Shapiro (2011), we do not consider this issue as a deadlock for his
comparative theory of justice. We rather view it as an invitation to reasoning.
As we have seen, in the TMS, one of Smith’s concerns is to explain how individuals’ judgments
are able to evolve toward more impartiality. This process involves the interaction between two
important figures: what Smith calls “the man without”, on the one hand, and the “man within”
(the genuine impartial spectator), on the other hand. According to us, the latter could constitute
the missing piece of Sen’s project. More precisely, we believe that Smith’s impartial spectator
could be extending to Sen’s analysis. For instance, it is possible to take inspiration from the
Scottish philosopher’s to represent internalized common agreed beliefs internalized by people
from the focal group and shaped through their social interactions. Such an extension of Smith’s
impartial spectator could supplement what we can now reasonably call Sen’s “man without” in
the process of public reasoning which must lead collective choices toward more justice.
In this sense, we believe that the impartial spectator serving Smith’s project could really serve
Sen’s own project.
4. References
Baujard, A., Gilardone M. & Salles M. (Forthcoming). A Conversation with Amartya Sen. Social
Choice and Welfare (Based on the 2010 interview in Harvard University).
Bréban, L. (2014), “Smith on Happiness: Towards a Gravitational Theory”, European Journal of
the History of Economic Thought, 21(3), pp. 359-391.
Bruni, L. (2011), “The Idea of Justice, Amartya Sen, The Belknap Press of Harvard University
Press, 2009, 467 pp.”, Economics and Philosophy, 27 (03): 324-331.
Campbell, T.D (1971), Adam Smith’s Science of Morals, London : George Allen and Unwin.
Clare, J. and Horn, T. (2010), “Sen and Sensibility”, South African Journal of Philosophy, 29 (2): 7484.
Dellemotte, J. (2011). La cohérence d’Adam Smith, problèmes et solutions : une synthèse critique
de la littérature après 1976, Economies et Sociétés, série Histoire de la pensée
économique, 45, pp. 2227-2265.
Dworkin, R. (2000), Sovereign Virtue: The Theory and Practice of Equality, Cambridge, MA:
Harvard University Press.
Ege, R. & Igersheim, H. & Le Chapelain, C. (2013). Par delà le transcendantal et le comparatif :
Deux arguments, Revue Française d’Economie, 27(4): 185-212.
Ege, R. & Igersheim, H. & Le Chapelain, C. (Forthcoming), « Transcendental vs Comparative
Approaches to Justice: A Reappraisal of Sen’s dichotomy”, The European Journal of the
History of Economic Thought.
Evensky, J. (1987). The Two Voices of Adam Smith: Moral Philosopher and Social Critic, History
of Political Economy, 19(3), pp. 447-468.
Fleischacker, S. (2011), “Adam Smith and Cultural Relativism”, Erasmus Journal for Philosophy
and Economics, 4(2): 20-41.
16
Forman-Barzilai, F. (2010), Adam Smith and the Circle of Sympathy, Cambridge: Cambridge
University Press.
Gauthier, D. (1986), Morals by Agreement, Oxford: Clarendon Press.
Gilardone, M. (2010), “A Review of “Amartya K. Sen (2009), The Idea of Justice, Harvard
University Press””, Social Choice and Welfare, 35(4): 709-720.
Gilardone, M. (2015), “Rawls’s influence and counter-influence on Sen: Post-welfarism and
impartiality”, The European Journal of the History of Economic Thought, 22(2):
forthcoming.
Hanley, R.P (2009), Adam Smith and the Character of Virtue, Cambridge: Cambridge University
Press.
Kandil, F. (2010), « Idéale ou comparative : quelle approche pour la justice sociale ? », Revue
Economique, 61 : 213-235.
Heilbroner, R. (1982). The Socialization of the Individual in Adam Smith, History of Political
Economy, 14(3), pp. 427-439.
Keppler, J-H. (2010), Adam Smith and the Economy of the Passions, Routledge.
Morrow, G. R. (1923). The Significance of the Doctrine of Sympathy in Hume and Adam Smith,
The Philosophical Review, 32, pp. 60-78.
Nozick, R. (1974), Anarchy, State and Utopia, New York: Basic Books.
Raphael, D. D. and Macfie, A. L. (1976), Introduction to the Theory of Moral Sentiments, in The
Theory of Moral Sentiments [eds. By David D. Raphael, Alec L. Macfie], Oxford: Clarendon
Press, 1976.
Rawls, J. (1971). A Theory of Justice, Cambridge Massachusetts: The Belknap Press of Harvard
University Press.
Robeyns, I. (2012), “Are transcendental theories of justice redundant?”, Journal of Economic
Methodology, 19 (2), 159-163.
Sen, A.K. (2002). Open and closed impartiality, The Journal of Philosophy, 99(9): 445–469.
Sen, A.K. (2009). The Idea of Justice, Cambridge Massachusetts: The Belknap Press of Harvard
University Press.
Smith, A. (1982[1759, 1790]), The Theory of Moral Sentiments, Raphael D.D. and Macfie A.L. (eds)
Oxford: Oxford University Press.
17