TRAGIC RECOGNITION: ACTION AND IDENTITY IN ANTIGONE ...

TRAGIC RECOGNITION: ACTION AND IDENTITY
IN ANTIGONE AND ARISTOTLE
Patchen Markell
The University of Chicago
Department of Political Science
5828 S. University Ave.
Chicago, IL 60637
p-markell@uchicago.edu
Prepared for the Center for Law, Culture, and Social Thought,
Northwestern University, March 9, 2001.
Comments welcome. Not for citation or circulation without permission.

This unchangeable identity of the person, though disclosing itself
intangibly in act and speech, becomes tangible only in the story of
the actor’s and speaker’s life; but as such it can be known, that
is, grasped as a palpable entity only after it has come to its end.
1. The politics and poetics of recognition.
1
Hannah Arendt 1
In a famous passage in the Letter to D’Alembert, Rousseau proposed a form of
“entertainment” to replace the shadowy corruption of the theatre: “Plant a stake crowned with
flowers in the middle of a square; gather the people together there, and you will have a festival.
Do better yet; let the spectators become an entertainment to themselves; make them actors
themselves; do it so that each sees and loves himself in the others so that all will be better
united.” 2 This is a vision of a society bound together, happily, by mutual recognition.
Contemporary political philosophers have cited it as an exemplary early statement of the now-
familiar ideal of equal and reciprocal recognition: just as Rousseau’s republican citizens exchanged
a love grounded in unimpeded mutual knowledge, so the contemporary discourse of recognition
envisions a world in which citizens extend each other forms of respect grounded in the accurate,
undistorted mutual cognition of their identities. 3

Of course, the tradition of dramatic theatre that Rousseau scorned was no stranger to the
notion of recognition. Against the background of the tradition of “recognition-scenes” in tragic
drama—Oedipus’s shattering moment of self-discovery, for example, or Electra’s recognition of
Orestes via his footprints and hair—Aristotle had introduced the concept of recognition,
anagnôrisis, to name one of the constitutive elements of a fine tragedy; 4 since then, recognition
has been both a crucial element in dramatic plots and a central concept in poetics. 5 On the tragic
stage, however, recognition looks startlingly different. In Rousseau as in contemporary political
theory, successful or adequate recognition is presented as a source of satisfaction, fulfillment, and
emancipation; tragic recognition-scenes, by contrast, are often moments of catastrophic loss and
occasions for mourning. 6
What lies behind this striking difference in valence? What can theories and examples of
tragic anagnôrisis teach contemporary political theory about the concept of recognition? This
paper explores these questions through readings of Sophocles’ Antigone and Aristotle’s Poetics.
In the larger work of which this paper is a part, I argue that contemporary theories and practices
of recognition are grounded in more fundamental, “ontological” misrecognitions—that is,
misrecognitions not of this or that identity, but of certain fundamental features of the social and
political world and our place in it. 7 The Antigone, I suggest, can help us identify and understand
these misrecognitions, because it depicts characters who put into practice some of the same
ontological assumptions that lie behind the contemporary discourse of recognition, and in so
doing, suffer tragic fates. Moreover, since tragedy is not just action but the representation of
action, 8 it keeps its audience or readers a step removed from the immediacy of practice, enabling
us to “see a whole denied to those who enact their parts,” and thereby helping us reflect on the
2

nature, sources, and consequences of the tragic characters’ blind spots—which may also, in other
contexts, be ours. 9
The particular ontological issue that concerns me here is the relationship between identity
and action, although, as I shall suggest, this issue is also tied to larger questions about sovereignty
and finitude. What the central characters in the Antigone share with the contemporary discourse
of recognition, I claim, is a fundamental faith in the capacity of “identities”—regarded in this case
as coherent systems of value and affect, anchored in specific locations in the human world—to
serve as the prior ground of agency, reliably guiding and governing action. This understanding of
the connection between identity and action is especially vivid in contemporary discussions of
multiculturalism. Some theorists argue straightforwardly that the cultural identity of a group
“defines” or “determines” its members’ activities, while others contend that the constitutive
relationship between identity and agency is mediated by choice—but the basic idea remains the
same: we know what we are to do, at least in some significant part, by knowing who we are; and
this is why inadequate or improper recognition of cultural identities can (it’s claimed) “cripple”
those who are subject to it. 10
At one level, of course, this view of the relationship of action to identity, by situating our
agency in social and historical context, reflects an admirable skepticism toward what has
famously been called the idea of the “unencumbered subject.” 11 But to say that subjects are
situated in and constitutively encumbered by contexts and histories beyond their control does not
entail that those encumbrances form coherent, intelligible wholes that determine our action in
absolutely predictable ways, and which could thus be the objects of truthful, and therefore
liberating, acts of recognition. Indeed, that idea too easily sidesteps the most radical implications
of the critique of the unsituated self. Even as it attacks the fantasy of a subject always already at
3

one with itself by virtue of its transcendence of context and history, the discourse of recognition
salvages the wish for sovereign agency that lies at the heart of that fantasy, envisioning a subject
restored to itself through the accurate and respectful cognition of its identity—a subject, in other
words, that wins its sovereignty in the thick of context and history, rather than above it.
As I have said, the antagonists in the Antigone share this view of the relationship between
action and identity, as well as the aspiration to sovereignty that underlies it. Yet this is only
Sophocles’ point of departure: the movement of the tragedy cuts against this view. The next two
sections of this paper trace this movement: in section two, I spell out the idea that the Antigone
can be read as a paradigmatic struggle for recognition, structured by the assumption of the
priority of identity to action; in section 3, I draw out an implicit critique of this assumption in
the action of the play, tracking the ways in which Antigone’s and Creon’s acts exceed the terms
of identity by which they seek to govern their conduct. I call this feature of action its
“impropriety” with respect to identity. In section 4, I turn to Aristotle’s Poetics to develop a
more explicit account of the nature, sources, and consequences of action’s improprieties, tying
the Antigone’s story about the relationship of action to identity to Aristotle’s account of the
relationship of action to êthos or character. 12 And these connections return us, in the end, to the
concept of recognition in Aristotle’s Poetics, as well as to the distinctive recognition-scenes of
the Antigone. The distinctive meaning and valence of “recognition” in tragedy, I shall suggest,
arises out of its unorthodox perspective on the relationship between action and identity: tragic
anagnôrisis involves not getting identity right so that it can ground our action; instead, it involves
acknowledging, often under the weight of failure, the limits to the possibility of doing so.
2. The Antigone as a struggle for recognition.
4

Sophocles’ Antigone is set after the death of Oedipus and in the wake of the deadly
rivalry between his sons, Polyneices and Eteocles. 13 After being exiled by his brother,
Polyneices led an Argive army against Thebes, and the two brothers died at each other’s hands. 14
Since the only surviving descendants of Oedipus were the sisters Antigone and Ismene, the
kingship of Thebes fell to Jocasta’s brother Creon, who faced the task of deciding how to treat
the bodies of the fallen warriors. 15 Creon gives Eteocles a hero’s burial, but orders Polyneices’
corpse left unburied and unlamented, and declares that “anyone who dares attempt the act” of
burial “will die by public stoning in the town” (35–36). As the play opens, Antigone tells
Ismene of Creon’s decree, and asks her to help defy the command and bury their brother’s body.
When Ismene refuses and reminds Antigone of her obligation to obey the city’s leaders, Antigone
does the deed herself. The outraged Creon condemns both sisters to death. Creon’s son
Haemon, to whom Antigone is betrothed, protests his father’s judgment, but to little avail.
Creon relents and releases Ismene, but orders Antigone sealed alive in a cave outside the city.
After Antigone, lamenting her fate, has been led away, Creon is visited by the prophet Teiresias,
who foretells that Creon’s impious acts will be repaid by death in Creon’s own family, and will
bring destruction to Thebes. Frightened, Creon tries to reverse his actions: he rushes to bury
Polyneices’ corpse and then to free Antigone. But Antigone has already hung herself, and
Haemon has already discovered her body. Creon arrives at Antigone’s tomb just in time to
witness his son’s angry, desperate suicide; and when news of Haemon’s death reaches his
mother, Eurydice, she too dies by her own hand, cursing Creon with her final breath. As the play
draws to a close, Creon, acknowledging his guilt, prays for an early death.
5

At first glance, the Antigone doesn’t seem to contain anything like the archetypal
“recognition scenes” of Greek literature: Odysseus gradually revealing himself after returning to
Ithaca in disguise; Oedipus discovering that he himself is the murderer whose identity he has
sworn to discover; Electra finding traces of and finally recognizing Orestes at Agamemnon’s
tomb. 16 Yet even without telltale scars and familiar locks of hair, the Antigone is full of
recognitions. The conflict with which the play begins is essentially a controversy about how to
recognize Polyneices; moreover, that controversy involves two adversaries who ground their
actions in confident claims about who Polyneices is, and also, as we shall see, in claims about
their own identities. In this way, the opening of the Antigone models the theoretical
underpinnings of the contemporary politics of recognition: the antagonists attempt to ground
their own actions in their knowledge of identities, and, on the basis of these properly cognized
identities, to secure respect from others. 17
The question of recognition first arises with respect to the dead Polyneices, whose body
is the principal object of controversy in the Antigone. The idea of a conflict over the burial of
bodies was probably not original with Sophocles. Attic funeral orators often recalled the story of
Thebes’ refusal to allow the burial of the corpses of the invading Argive army, which supposedly
provoked a heroic Athenian expedition to recover the bodies. 18 But these were simple, self-
congratulatory stories of Athenian virtue in the face of sheer impiety, and not the stuff of deep
ethical conflict. 19 Sophocles’ version of the story focuses our attention on problems of identity
and recognition by zeroing in on the disparate treatment of Eteocles and Polyneices. 20 The case
of Polyneices and Eteocles is distinctive, because the question of how to deal with Polyneices’
body seems to hangs on who he is—and he can be recognized under two incompatible
descriptions, each of which highlights one aspect of his relationship to Eteocles. On the one
6

hand, Polyneices is a native of Thebes, a member of the ruling family, the brother of Antigone
and Ismene, and the nephew of Creon himself—all relations that throw his affiliation with
Eteocles into relief. Under this description, Polyneices unquestionably ought to be buried, for it
was not only a right but an obligation to give the proper funeral rites to one’s kin. 21 On the other
hand, Polyneices is a traitor who has just raised an army against Thebes out of jealousy and
vindictiveness toward his brother. 22 This perspective makes the difference between Polyneices
and Eteocles more salient than the fact of their affiliation, and demands that they be treated
differently in death. 23
Here, then, as in the contemporary politics of recognition, cognition and respect are
brought together in a distinctive constellation. The distribution of respect is supposed to be
grounded in knowledge of identity; thus, the disagreement about whether Polyneices is a kinsman
or a traitor engenders a further conflict about whether he ought to be granted a funeral, or left to
rot. Indeed, the theme of burial illustrates the close relationship between these two components
of recognition—cognition and respect—with exceptional clarity. Funeral rites are an expression
of respect: they affirm the membership of the deceased in a cross-generational community, and
they create monuments to the dead that help to preserve traces of this community within the
world of the living. Moreover, funeral rites confer honor and affirmation precisely through the
articulation of the identity of the deceased. The spontaneous lament, the public funeral oration
and the carefully composed elegy all recollect the dead person’s character and actions, and the
epitaph carved on the funeral monument helps to “secure a degree of personal immortality” for
the deceased by delivering at least his name, and sometimes a record of his deeds and virtues, to
posterity. 24 In burial, as in the contemporary politics of recognition, the public visibility of
identity becomes an instrument for securing respect.
7

Yet the Antigone’s struggle is not only about recognizing and respecting Polyneices.
Although his death is the immediate occasion for the conflict between Antigone and Creon, the
significance of their confrontation quickly widens, for both ground their obligations toward the
corpse not only in their descriptions of Polyneices but also in accounts of their own identities. If
we attend again to the significance of the themes of burial and lament in the Antigone, it should
come as no surprise that the identities of Antigone and Creon are drawn into the controversy
over Polyneices, for funeral ritual is at least as much about the living as about the dead. As
psychological studies of mourning emphasize, the same rites of burial and lament that serve to
memorialize the dead also help to reconstitute the agency of the mourner in the face of loss. 25
The linguistic and physical artifacts produced as a part of funeral ritual—laments, eulogies, grave
markers—cannot really replace the lost person. 26 But in the production of such artifacts, and
through participation in the symbolic systems that govern funeral procedure, mourners try to
reconstruct the agency that the intrusive facticity of death had seemed to undermine. 27 Of
course, that agency is never experienced immediately and abstractly, but only by occupying
particular roles and identities in the social world; and since the roles and identities through which
different people experience agency may make incompatible demands in a single case, an instance
of mourning can quickly become the occasion for broader social conflict. In this way, the struggle
over Polyneices turns into a struggle for recognition between Antigone and Creon, as they try
unsuccessfully to compel each other to acknowledge not just some facets of Polyneices’ identity,
but the legitimacy of their own identities—of the locations in the ethical world from which
Antigone and Creon try to address the personal and civic losses Polyneices’ body represents.
Antigone and Creon introduce themselves, and articulate the identities in which they
ground their acts, in the first two episodes of the play: Antigone’s exchange with Ismene, and
8

Creon’s speech to the chorus. These scenes show the audience for the first time who these
characters are, and they do so by locating them within the social world that constitutes the
background to the events of the play. As Simon Goldhill and others have observed, one of the
most important ways Antigone and Creon locate themselves is via their contrasting uses of the
opposition between philos and ekhthros, which both characters employ with reference to
Polyneices. 28 As a concrete noun, philos is usually translated “friend”; as an adjective, the same
word is rendered as “dear,” “beloved” or “loving”; and the related noun philia is translated as
“friendship” or “love.” Correspondingly, ekhthros, as a concrete noun, is translated “enemy”; as
an adjective it is rendered as “hostile” or “hated”; and the abstract noun ekhthra is translated as
“hatred” or “enmity.” 29 Yet these translations do not do justice to the semantic range of the
words, especially in the case of philos and philia, which could be used in the context of nearly
any “positive” reciprocal relationships, including bonds among kin, strategic alliances, the
extension of hospitality to strangers, self-love, and the ties of marriage, as well as the class of
personal relationships of mutual affection and support that we normally call “friendship.” 30
Precisely because of their semantic breadth, these words are well-suited to become the focus of a
struggle over the legitimacy and priority of different kinds of social bond. Antigone and Creon
make good use of this potential, transforming what is initially a debate about Polyneices into a
conflict over the proper sense of philos and ekhthros—and thereby announcing their own deepest
commitments to the audience. 31
Antigone’s exchange with Ismene, for instance, is framed by the theme of philia. In the
opening lines of the play, Antigone approaches Ismene with news of Creon’s edict: “Have you
heard anything?” she asks, “or don’t you know that the foes’ [ekhthrôn] trouble comes upon our
friends [philous]” (10)? Antigone implies that in denying burial not only to the six fallen Argive
9

commanders but also to Polyneices, Creon has brought an evil appropriate to an ekhthros upon a
philos. 32 We do not yet know in virtue of what characteristic or relationship Antigone considers
Polyneices philos, but in the course of her exchange with Ismene, Antigone’s meaning becomes
clear.
First, Antigone indicates that she considers Polyneices philos by virtue of kinship. When
Ismene expresses her surprise at Antigone’s plan of disobedience, Antigone justifies herself
simply by referring to the fact of family ties. Polyneices is “my brother, and yours, though you
may wish he were not,” Antigone says. In this declaration, she also identifies herself by placing
herself in a network of blood ties and expressing her allegiance to the bonds of the family.
Second, Antigone indicates that her obligations of philia to Polyneices arise from what she will
later call the “unwritten and unfailing laws” of “the gods below” (451, 455), which govern
reproduction and death. 33 Ismene introduces the theme, contrasting “them below the earth” with
“the men in power” in the city—but while Ismene declares that she feels compelled to obey the
city (65–67), Antigone responds that she regards the underworld as of greater import than the
world of the living, and suggests that once she has “dared the crime of piety,” she will be able to
lie alongside Polyneices, philos with philos (73–74). Finally, although the gendering of the
conflict between Antigone and Creon is more explicit in Creon’s speeches, Antigone does some
of the work of gender identification in her exchange with Ismene. Although the contrasting case
of Ismene, who submissively cautions her sister to “remember that we two are women” who
should not fight with men (61–62), might be taken to be indirect evidence of Antigone’s
resistance to any identification with conventional gender roles, the truth is more complex. As
Charles Segal has explained, for instance, the vocabulary of kinship with which Antigone refers to
her siblings is itself gendered: it “makes kinship a function of the female procreative power,” in
10

direct contrast to the patrilineal system of kinship on which membership in the polis was based,
and from which women were excluded. The ordinary words for brother and sister, adelphos and
adelphê, literally indicate kinship through the womb (delphus), a variety of kinship to which
Antigone calls attention when she later calls Polyneices homosplanchnos—a less ordinary word
that also means “of the same womb” (511). 34 Should the etymologies seem trivial, recall that the
question of the nature of kinship and its relation to gender was a live issue for the Greek
audience: as Segal points out, Antigone’s use of matrilineal kinship words “reopens, on a
personal level, the debate between Apollo and the Erinyes in Aeschylus’s Oresteia,” where the
issue was precisely whether the father’s seed or the mother’s womb played a more important
role in reproduction. 35
Moreover, Antigone’s identification with (at least some) conventional gender roles is later
borne out by the manner in which she performs her duties of philia to Polyneices. Although she
does initially declare her intention to “pile the burial-mound” for Polyneices (80)—a task that
was generally performed by men—she does not actually bury the body but merely sprinkles dust
on the corpse, “enough to turn the curse,” as the guard says (255). 36 Later, just before being led
away by Creon’s guards, she describes her action in a way that conforms quite precisely to
classical norms governing the role of women in funeral rites: “All three of you have known my
hand in death,” she says, referring to Oedipus, Jocasta, and Polyneices; “I washed your bodies,
dressed them for the grave, poured out the last libation at the tomb” (900–902). And, finally,
Sophocles describes Antigone’s lament over Polyneices’ corpse with a familiar, gendered image
specifically associated with the mourning of mothers for their children: “the sharp and shrill cry
of a bitter bird which sees the nest bare where the young birds lay” (423–24). 37
11

Creon’s first speech sharply contests Antigone’s understanding of philia. Whereas
Antigone regards Polyneices as philos by virtue of kinship, treats her own obligations to him as a
matter of chthonian piety, and expresses a gender identity in the way she grounds and performs
these obligations, Creon explicitly rejects each of these identifications. First, although Polyneices
is Creon’s kinsman too, Creon treats him as ekhthros, not philos, because for Creon the criteria of
philia are exclusively political. “He who counts another greater friend [philon] than his own
fatherland [patra], I put him nowhere,” Creon announces (182–83), already invoking the
principle of patrilineal descent that underlay membership in the polis against Antigone’s
matrilineal vocabulary of kinship. Only the ship of state, “sailing straight,” can make it possible
for us to “have friends [philous] at all,” Creon asserts (189–90). For Creon, Polyneices ceased to
be a philos the moment he raised an army against Thebes, and after she violates the city’s laws,
Antigone too will be called a “false friend” [philos kakos] (652). 38 In identifying himself instead
as a citizen and ruler whose deepest commitment is to the all-important law of the polis, Creon
trumps both the principle of kinship and the eternal and unwritten chthonian law that
supposedly assigns duties within the family.
Creon’s exclusively civic conception of philia is also rigidly masculine. As Creon’s
polemical use of the word patra as a synonym for the polis has already indicated, citizenship and
rulership are properly the business of men, and only men. The intertwining of gender identity
with the distinction between polis and oikos, already implied in Creon’s first speech, runs so
deep that Creon, upon hearing the guard’s report that Polyneices’ corpse has been buried, thinks
only to ask “what man [tis andrôn] has dared to do it” (248). The fact that Antigone has
disobeyed thus turns out to be an especially potent threat to Creon’s authority: it throws into
question not just his authority in this particular case, but his masculinity, the condition of the
12

possibility of his political authority as such. In response, Creon’s gendering of the polis
becomes more explicit, insistent, even hysterical: he repeats in different ways and to various
people that he will never let himself be conquered, ruled, defeated by a woman. 39 In one
remarkable passage Creon ties together his renunciation of the ties of kinship as grounds of philia
and the preservation of his own masculinity:
I am no man and she the man instead
if she can have this conquest without pain.
She is my sister’s [adelphês] child, but were she child
of closer kin than any at my hearth,
she and her sister should not so escape
their death and doom (484–89).
These, then, are Antigone’s and Creon’s representations of their own identities (and Polyneices’)
which they deploy in their efforts to guide their own actions and to influence or control the
actions of others. Their confident self-descriptions fit well with the common view that Antigone
represents the family, the chthonian, and the female, while Creon represents the city, the human,
and the male. 40 Yet by focusing on what Antigone and Creon properly represent, we would
ignore the movement and action of the play. For when Antigone and Creon try to fulfill or enact
their identities in their deeds, they find that their actions do not flow smoothly from their
identities: as we shall see, they commit what I shall call “improprieties” with respect to the
identities they claim to bear, and these improprieties place them at odds with their own deepest
commitments. The fact that Antigone and Creon’s deeds cannot easily be reconciled with (what
they take to be) their identities is not evidence of a failure in Sophocles’ characterization; rather,
it is a comment about the nature of identity, and about the limits of our capacity to deploy
identity to predict and govern the meanings and consequences of our actions.
3. The improprieties of action.
13

What happens when Antigone and Creon attempt to ground their actions in their
identities? Consider Antigone first. Her action exceeds and frustrates her identifications in two
related ways: first, Antigone does more than she intends. Antigone frames her action as an
expression of the pious devotion of sister to brother, and she underscores this identification
through her refusal of the vocabulary of politics. Whenever Antigone is confronted with a claim
about the city, she replies exclusively in the vocabulary of the family, refusing the possibility of
any distinction between her brothers, even one more attenuated than Creon’s. As Martha
Nussbaum puts it, “if one listened only to Antigone, one would not know that a war had taken
place or that anything called ‘city’ was ever in danger.” 41 Despite her refusal of the vocabulary
of politics, however, her expression of family piety turns out also to be an act of political
subversion.
However, Antigone does not merely turn out to have done more than she had intended,
for her act places her into conflict not only with what she disavows but also with her own
deepest commitments. Antigone’s relationship with Ismene is a telling example of this second
kind of impropriety. Although Antigone is willing to suffer death out of loyalty to a blood-
relative, in the pursuit of her goal she behaves toward her sister—her adelphê, born of the same
womb just like Polyneices—with cold, vindictive hostility. When Ismene tries to warn Antigone
that it is foolish to pursue her goal against the irresistible force of the polis, Antigone responds:
“If that’s your saying, I shall hate you first, and next the dead will hate you in all justice”
(93–94). And when Ismene shows her belated support for her sister by falsely declaring to
Creon that she had been an accessory to the deed, Antigone declines the offer of solidarity,
insisting that she “cannot love a friend whose love is words” (543)—and this comes mere
14

moments after Antigone declares to Creon that it is her nature to love, not to hate (523)! 42 In
pursuit of divine justice for her brother, Antigone undermines her own identification with the
principle of devotion to the family by declaring her hatred for the sister whom, on her own
account, she ought to love.
Antigone’s action subverts not only her identification as a member of a family, but also
her identification with conventional gender roles. As we have seen, by preparing Polyneices’
body for burial, scattering dust on the corpse to symbolize the burial that she herself does not
perform, lamenting Polyneices’ death, and pouring a final libation, Antigone follows the norms
governing female participation in funeral ritual. But Antigone acts amid circumstances that make
it impossible for her to conform to this traditional role without also violating equally central
norms of gender.
Antigone’s impropriety with respect to conventional gender roles is most clearly evident
in Creon’s hysterical response to her disobedience, which reminds us that the very appearance of
women in civic space threatened the constitutive boundary between oikos and polis. Thus, for
example, after pronouncing his initial sentence on Antigone and Ismene, Creon orders his slaves
to take them back inside the walls of the palace, insisting: “They must be women now. No more
free running” (578–79). 43 The form of punishment Creon finally chooses for Antigone is an even
more radical enclosure of this improper woman: Antigone herself notices the analogy between the
cave in which she is to be buried and the oikos in which women are confined, for she calls the
cave her “marriage-chamber” and her “hollowed-out house” (891–92). 44 Indeed, although Creon
himself never indicates such a motivation, it is noteworthy that his strategy of burying Antigone
alive rather than having her stoned to death, as originally planned, conveniently sidesteps the
issue of Antigone’s funeral rites: confined to the cave, she will die “with no friend’s mourning”
15

(844). Given the function of the lament and the epitaph in conferring worldly immortality,
perhaps the punishment of an unwitnessed death was intended to prevent Antigone and her acts
of resistance from being memorialized in the city; to keep even her name out of the public
sphere—which would, again, restore a conventional order of gender in which only men were
properly bearers of individual distinction, deserving of public remembrance. 45
Yet we do not need to assume the perspective of the anxiously masculine Creon in order
to understand Antigone’s action as an impropriety with respect to gender identity. Antigone’s
action frustrates her own identification as a woman even from her own perspective, by blocking
the usual trajectory from girlhood to marriage and reproduction—and thereby also shattering her
identification with her mother. In an important recent essay that touches on these themes, Matt
Neuburg shows the importance of the distinction between blood-ties and marriage-ties in the
Antigone. “Antigone will never get to have a marriage-family,” Neuburg observes, because “she is
going to die for the sake of her blood-family: her two roles in the world, daughter and sister on
the one hand, wife and mother on the other, have come into apparently irreconcilable conflict.” 46
And indeed, Antigone’s own lament upon her fate reflects this conflict: “My curse,” she mourns,
“is to die unwed” (869). What Neuburg overlooks is that, because marriage-ties and blood-ties
are not only distinct but interdependent, Antigone’s deed also undermines her identification with
the principle of kinship through blood by depriving her of the opportunity to bear children: as
Antigone herself says, “no marriage-bed, no marriage-song for me, and since no wedding, so no
child to rear” (917–18, emphasis added). 47
Creon’s deeds, too, make him into an improper representative of the identities he initially
claimed as his own. As we have seen, Antigone’s act takes on a political meaning to which she
remains obstinately blind in her commitment to the family and its law. Likewise, Creon’s act
16

turns out to have familial implications that he, focused exclusively on the polis, does not
acknowledge. Creon is, after all, not only the ruler of Thebes but also the head of the oikos to
which Antigone belongs; indeed, he is doubly tied to her both as Jocasta’s brother and as the
father of her fiancé. But in his encounter with Antigone, Creon notably does not explicitly
invoke his familial authority; instead, he continues to assert the political distinction between
Polyneices and Eteocles, just as Antigone repeatedly counters him in the vocabulary of kinship.
When Haemon arrives on the scene, he invites his father to speak the language of family by
offering a conventional expression of filial loyalty: “My father, I am yours. You keep me
straight with your good judgment, which I shall ever follow” (635–36). 48 But Creon refuses the
invitation and quickly shifts the terms of the conversation, swallowing family into polis. He
replies by praising Haemon, but immediately introduces the metaphor of “a soldier posted
behind his leader” to describe the proper relation of son to father (640). 49 Finally, although
Creon nominally acknowledges that Antigone’s disobedience is an instance of disorder within the
oikos, he does so only in order to deny that Antigone’s kinship ought to influence his action, and
to reinforce his subordination of all other concerns to political rule. To permit disobedience
among relatives, Creon says, would compel him to permit it in the city at large; thus, enforcing
the edict against Antigone is just another instance of ensuring “justice in the polis” (662). But
Creon’s act, like Antigone’s, exceeds the identity from which it proceeds. Just as Antigone’s act
of family piety was also an act of political subversion, Creon’s defense of political order also
turns out to be an assault on his own family, first in the person of Haemon, whose love for
Antigone drives him to join her in death, and second in Eurydice, driven to suicide by the loss of
her son.
17

Worse still, Creon’s acts, like Antigone’s, not only exceed but also undermine his own
identifications. Just as Antigone’s ruthless devotion to her family leads her to treat her own
sister cruelly, for instance, Creon’s monomaniacal pursuit of civic order transforms him from a
leader into a tyrant. After Creon shifts the terrain of his exchange with Haemon from family to
city, Haemon criticizes his father’s style of rule, at first gently and then with increasing passion
and frustration. After reminding Creon of the murmurs of support for Antigone among the
citizens of Thebes, Haemon warns his father: “do not have one mind, and one alone,” for
“whoever thinks that he alone is wise” will, “come the unfolding, sho[w] his emptiness” (705–9).
Here already we have a pointed suggestion that Creon is betraying his own values, since he
himself had earlier mocked Antigone for acting in isolation from the rest of the city (510);
Teiresias will confirm the suggestion by insinuating that Creon is not a king but a “tyrant”
(1056), and by forecasting the devastation that Creon’s misrule will bring, not only upon his
family, but upon the entire city (1080–84).
Finally, just as Antigone’s action exceeds and frustrates her identification as a woman,
Creon’s deeds, though meant to secure his masculinity, actually undermine it. For Creon, as we
have seen, masculinity is closely tied to political rule, and therefore (he insists) to the rejection of
the normative force of kinship. But consanguinity was what brought him the political power
through which he expresses his manhood: he rules Thebes by virtue of being Jocasta’s
adelphos—the very same relation in which Polyneices and Antigone stand. His disavowal of the
import of kinship leads to the death of his own son, Haemon, in a violent rush of blood (haima).
As Creon is graphically reminded of the blood-ties he had repressed, his fragile economy of
identification and disavowal collapses like a house of cards. “So insistent earlier on the
separation of gender roles and scornful of the female,” Segal observes, “Creon now performs the
18

characteristically (though not exclusively) female role of lamenting over a ‘child,’” expressing the
destabilizing anguish and bitterness that his edict had tried and failed to keep out of public
view. 50
Perhaps the most powerful image that the Antigone offers of the capacity of actions to
exceed the intentions and control of agents comes near the end of the play, as Creon tries to undo
what he has done. After confronting Teiresias, who finally gives Creon an account of his crimes
and foretells the destruction that await him and his city, Creon yields, and tries to stave off his
fate, announcing that he has “come to fear it’s best to hold to the laws of old tradition to the end
of life” (1113–14). Some of what happens next we witness or learn through the reports of a
messenger before Creon does, but the special force of the sequence of events emerges if we
examine it in strict chronology and from Creon’s perspective. First, as the messenger recounts,
Creon and the chorus went to bury Polyneices’ corpse, hoping to reverse the pollution that the
exposed body had brought upon the city (1197–1200). 51 It’s not clear whether Creon relents in
time to turn the curse or not, since Teiresias has said that other cities, whose hearths have been
polluted by the animals that fed on the exposed corpses, are already preparing to make war on
Thebes (1080–84). 52 With no time to speculate, Creon and his rescue party proceed toward
Antigone’s tomb, from which, at a distance, they hear the “keen lament” of Haemon’s voice
(1208). Antigone has already hanged herself, and while Creon was busy burying Polyneices,
Haemon found the tomb and discovered her fate. When Creon finally reaches the burial chamber,
he finds Haemon mournfully embracing Antigone; Haemon says nothing to his father, but lashes
out at him with his sword, and then turns his weapon on himself, dying in a bloody embrace with
Antigone (1220–40). At this point, the messenger leaves Creon and the rest of the party to
attend to the bodies and rushes back to the palace, where he informs Eurydice (and us) of all the
19

preceding events; she leaves the room without a word, followed in short order by the messenger,
who is concerned that her silence “may portend as great disaster as a loud lament” (1251–52).
Creon arrives shortly thereafter—and just as he is confessing his crimes to the assembled chorus,
the messenger returns to announce that Eurydice, too, has taken her own life (1282–83).
Once Creon has given it life, we might say, his action breaks the bonds of intention and
identity and goes off on its own, wreaking havoc. Creon rushes from the house to the deserted
plain to the rocky cave and back to the house, all in a desperate attempt to chase down and
subdue his errant deed, but the act and its reverberations always remain one tragic step ahead of
its agent. In the case of the messenger’s early departure from the tomb, the deed, or at least its
“signification,” is literally a step ahead of the doer. 53 Here, the character of the messenger vividly
personifies the gap between the actor’s performance of a deed and the imitation or recollection of
the deed in which its significance is expressed, offering tragic confirmation of Hannah Arendt’s
observation that “nobody is the author or producer of his own life story.” 54
4. The acknowledgment of finitude: Aristotle and the Antigone.
In chapter 6 of the Poetics, Aristotle makes a famous claim about the relative importance
of the constituent parts of tragic drama: “The most important of the six [parts of tragedy],” he
says, “is the combination of the incidents of the story,” for “tragedy is essentially an imitation
not of persons but of action and life” (1450a15–17). 55 Thus, for Aristotle, “the first essential,
the life and soul, so to speak, of tragedy is the plot,” while “the characters [êthê] come second”
(1450a38–39); indeed, the characters are included for the sake of the action rather than the other
way around (1450a20–22). 56 Aristotle’s point here is sometimes taken to be wholly formal: it is
20

part of a treatise articulating guidelines to be followed by practitioners of the craft of poetry,
from which modern readers might also glean some insight about how to interpret works of
literature. 57 Yet in the course of elaborating this claim about the priority of action and plot to
character, Aristotle moves, momentarily, into a different register:
Tragedy is essentially an imitation not of persons but of action and life, of
happiness and misery. All human happiness or misery takes the form of action;
the end for which we live is a kind of activity, not a quality. Character gives us
qualities, but it is in our actions—what we do—that we are happy or the reverse.
In a play accordingly they do not act in order to portray the characters; they
include the characters for the sake of the action (1450a16–22, emphasis added). 58
Aristotle’s claims in the emphasized sentences refer not only to the representation of
action and character on the stage, but to action and character simpliciter. 59 As Martha Nussbaum
has suggested in an important discussion of the relationship between the Poetics and Aristotle’s
ethics, these passages suggest that for Aristotle, tragic drama’s formal characteristics are not
merely formal, but reflect and illuminate features of the wider world. 60 Following this lead, I now
turn to Aristotle’s Poetics in order to work out, in greater detail, the nature and sources of the
“improprieties” that characterize Antigone’s and Creon’s actions, as well as to illuminate the
specific meaning of tragic anagnôrisis or recognition. 61 As I have suggested, Sophocles’ text
implies a critique of the assumptions about the relationship of action to identity that informed
Antigone’s and Creon’s practical postures (and which are more explicitly articulated in the
contemporary normative discourse of recognition). Aristotle’s claim that action has a kind of
priority over character, I claim, can be read as a philosophical distillation of that critique.
Moreover, tying Aristotle’s claims about action and character to his discussion of anagnôrisis
later in the Poetics (and also back to the Antigone) will help show that tragic anagnôrisis consists
21

not in the agency-enhancing recognition of an identity but rather in the acknowledgment what
Hannah Arendt calls the “non-sovereign” character of human action. 62
* * *
In order to bring Aristotle’s claims about action and character to bear on the themes of
action and identity we have been discussing so far, however, we need to deal with a
terminological question: what is the relationship between Aristotelian êthos and the notion of
identity? After all, it might be objected that Aristotelian êthos refers not to the full range of
socially contingent characteristics we bear, but to our specifically moral or ethical qualities—that
is, to the virtues and vices Aristotle discusses in his ethical treatises. 63 What does character in
this sense have to do with the “identities” articulated by Antigone and Creon—much less with
the use of “identity” in contemporary politics, with its heavy emphasis on group membership
and ascriptive characteristics rather than the moral states of individual ethical agents?
We can address this question by briefly examining the meaning of êthos, both in broader
context and in Aristotle’s philosophy. In an illuminating discussion of the development of the
term êthos from Homer down to Aristotle, Charles Chamberlain has shown that êthos originally
referred to the “haunts” of an animal—that is, to the places where animals of a particular type
were typically or characteristically found. 64 The term later came to be extended to human beings
and transformed in a number of ways—but far from being used exclusively to refer to the vices
and virtues of individuals, it was also used to refer to “the peculiarities which people of a certain
polis acquire as a result of being brought up under its particular laws and customs”; Thucydides,
for example, has Pericles remind the Athenians that they were brought up in a certain êthos,
which Rex Warner renders as “way of life.” 65 Importantly, then, the line of continuity that
Chamberlain traces from “haunts” to “character” does not suggest some exclusive connection
22

etween êthos and virtue. Instead, it suggests that êthos is simply something like that which
gives rise to a distinctive and stable pattern of behavior on the part of things of a certain kind—a
pattern of behavior whose virtuousness or viciousness is one, but only one, interesting thing
about it. 66
Chamberlain’s broader account of the meaning of êthos fits well with Aristotle’s own use
of the term. In the Nicomachean Ethics, Aristotle does not define êthos as such. But in book 2
of the Eudemian Ethics, Aristotle says that êthos is a “quality” [poiotês] that we possess in
respect of our “capacities for affections” [dunameis tôn pathêmatôn] such as anger, fear, shame,
or desire; as well as in respect of the states or habits [hexeis] that determine how we experience
and respond to those affections (1220b5–20). 67 A quality, in turn, is a “differentia” [diaphora]
(Metaph. 5.14, 1020a33); it is simply “that in virtue of which people are said to be such and
such” (Cat. 8, 8b25). And because the states or habits that differentiate us are “lasting and firmly
established” (Cat. 8, 8b27), when we give accounts of the êthos of a person, we are saying
something about what we take to be that person’s persistent qualities with respect to affection,
rather than their momentary attributes. In sum, as Nancy Sherman says, Aristotelian character
“has to do with a person’s enduring traits; that is, with the attitudes, sensibilities, and beliefs that
affect how a person sees, acts, and indeed lives.” 68
What do these genealogies and definitions show? There are two possible conclusions to
be drawn, either of which would be adequate to establish a productive connection between the
themes of action and character in the Poetics, and the themes of action and identity in the
Antigone and elsewhere. The first conclusion is stronger. If Aristotelian êthos, far from being
limited in meaning to the various qualities of virtuousness or viciousness we possess, actually
includes the whole range of “enduring traits” that influence our behavior (arguably excluding
23

those relating to thought or intellect), then êthos is capacious enough to include “identity”—for
“identity,” as I have been using it here, refers not simply to a set of personal or social
characteristics, but to a set of such characteristics that have practical force. So, for example,
when Creon, through his speeches, demonstrates how heavily his sense of the meaning of philia
(and therefore his understanding of his obligations and responsibilities) has been shaped by his
status as citizen and king, it makes sense to say that he has shown us something of his êthos: he
has laid out the terms that govern his action. 69
The second conclusion is a bit weaker. Even if we remain skeptical that êthos can
encompass identity, there is nevertheless an important analogy to be drawn between character
and identity on the basis of the way each relates to action. The preceding discussion should at
least have brought into the foreground the fact that êthos—even êthos narrowly construed—is
typically taken to be prior to action in the same way that Antigone and Creon (and the
contemporary normative discourse of recognition) take identity to be prior to action. Our
qualities of character are stable and persistent attributes that help to determine the actions we
perform (NE 2.2, 1104a27; 3.5, 1114b25), and for this reason, as Sherman says, character helps
us explain “not merely why someone acted this way now, but why someone can be counted on
to act in certain ways”; it introduces predictability, “pattern,” and order into our doings. 70
Indeed, this basic view of êthos has been understandably central to contemporary neo-
Aristotelian revivals of virtue ethics, which focus on the underlying qualities, formed through
practice, habituation, and education, that dispose people to act well: in his account of Aristotle’s
ethics, for example, MacIntyre writes that “the immediate outcome of the exercise of a virtue is a
choice which issues in right action,” and again that “the virtues are precisely those qualities the
possession of which will enable an individual to achieve eudaimonia and the lack of which will
24

frustrate movement toward that telos.” 71 And this, ultimately, is what enables Aristotle’s
consideration of character and action in the Poetics to shed light on the relationship between
identity and action, both in the Antigone and in general.
* * *
We began this section with Aristotle’s claim, in chapter 6 of the Poetics, that tragedy is an
imitation of action rather than of persons, and that in tragedy the plot and its constituent actions
have a certain kind of priority over character. But after the account of the meaning of êthos we
have just developed, this claim might seem puzzling: how can action have priority over character
if character is that set of stable qualities that governs and determines our actions? On some
accounts, this apparent contradiction can only be addressed by positing a deep gulf between
Aristotle’s action-centered account of tragic drama in the Poetics and his character-centered
undestanding of “real life” in his ethical works; if this is right, then it can only be misleading to
look to the Poetics for insight into the real-life relationship between character and action. 72
Such a conclusion might be inevitable if Aristotle’s account of the relationship between
character and action in “real life” were reducible to the thought that êthos determines action. But
important parts of the Nicomachean Ethics, including the discussions of habituation and the
acquisition of virtue in chapter 2, suggest that Aristotle holds a more complex view. It’s not only
that the virtue of bravery causes or is actualized in courageous acts, for example, but also that
bravery itself is formed through the doing of brave deeds (NE 2.2, 1104a28–b4); this, Aristotle
maintains, “is why we must perform the right activities” (NE 2.1, 1103b23). 73 For Aristotle,
then, action and character form a circuit of mutual constitution: character pushes out into the
world by shaping the kinds of things we do; and it is at the same time formed and reformed by
our worldly activity. Against this background, it becomes possible to find in the Poetics an
25

argument that is consistent with, though perhaps different in emphasis from, Aristotle’s broader
ethical views. The point of the Poetics’ claim about the priority of action over character, we
might say, is that this circuit of mutual constitution is not closed. While character influences
action, it does not perfectly determine it; and this means both that “being of a certain good human
character [may fall] short of sufficiency for eudaimonia,” and that character itself is
constitutively vulnerable to the various forms of worldly contingency that are transmitted,
through our activity, back to our being. 74
What are these “various forms of worldly contingency” that afflict activity? And how do
they relate to what, in the context of the Antigone, I have called action’s “impropriety” with
respect to identity? 75 The first question of these questions is, of course, enormous; I will merely
mention two very general considerations here. First, action projects human beings into a world of
causality, initiating sequences of events that, once begun, proceed without necessarily respecting
the agent’s intentions. This fact of the causality of human action most obviously threatens our
capacity to control the consequences of our actions. But it also limits our capacity to control the
significance of our actions, insofar as it prevents us from locating a natural and uncontroversial
boundary between our actions and the events that follow from them. To be sure, we rely on such
boundaries all the time, particularly in the law, where we often need to decide whether an event is
to be imputed to an agent as his act, for which he may be held responsible. But although these
conceptions of the limits of responsibility or imputability may represent themselves as
reflections of some sheerly factual line of demarcation between, say, the willed and the caused, it
is in a way the very absence of such a line that allows imputability to arise as a problem in the
first place. Will and agency only become possible sources of injury or damage because they are
not isolated from a separate world of causes and effects, but are themselves sources of
26

causation 76 —and this absence of a firm qualitative distinction between will and causation makes
it difficult to fix, once and for all, the limits of what may be imputed to us as “our” doing. 77
The fact that our action inserts us into chains of causality not wholly under our control
can, of course, manifest itself in numerous ways, and is perhaps most strikingly visible in cases
of natural disaster in which nonhuman forces undermine our plans (and often destroy us
altogether) in unpredictable, perhaps even utterly meaningless ways. 78 Still, the sheer
contingency of nature is not quite the stuff of tragedy—the abuse of the term “tragedy” in the
modern media notwithstanding. 79 A bit closer to tragedy’s concerns lies the fact that human
beings act into a world inhabited by a plurality of other acting persons: the fact of human
freedom, which is the condition of the possibility of the effective agency, also limits our practical
capacities because it is not exclusively ours but is mirrored in others. 80 At one level, of course,
this reference to the fact of human plurality can be subsumed under the point I have just made
about causality: one of the central reasons we cannot perfectly determine the consequences of our
actions is that we never act altogether in isolation but in interaction both with sometimes
uncooperative or intransigent others as well as with powerful, large-scale human institutions.
But the point is not merely that human plurality limits our control over the consequences of our
actions; here, too, it’s also that the meaning of our deeds is not wholly at our disposal, for the
very terms through which we make assessments of significance are not exclusively our own, but
intersubjective—as Creon and Antigone discover in their clash over the sense of philos and
ekhthros. 81
Notice that these general features of action share a distinctive structure: they involve the
doubling back of some human capacity upon itself—a recursivity, in which a source of
possibility also operates as its own limitation. If human beings were not themselves parts of the
27

causal order of the world, the will would be impotent; yet the very fact of our implication in the
causal order of the world both connects our deeds to chains of events that lie beyond our control,
and blurs the boundaries between action and event that might help us fix and master the meaning
of our actions. Likewise, the idea of freedom depends upon the human capacity to
spontaneously initiate the new and the unexpected—and yet the fact of human freedom inhibits
our ability to do as we will, since the new and unexpected actions of others may throw
unforeseen obstacles in our path. This peculiar structure of enabling conditions that are always
also limitations places human actors in an ontological double bind, rendering us dependent on the
very forces that, in action, we seek to transcend.
The instances of “impropriety” that we have observed in the action of the Antigone, I
suggest, can be understood as consequences of analogous double binds. 82 Antigone and Creon
work to recognize and be recognized in certain ways in part as response to the shattering losses
they have experienced: for them, the pursuit of recognition promises a satisfying restoration of a
sense of agency and independence in the face of the unpredictability and capriciousness of the
world. The categories of identity they claim and assert make it possible for them to know what
to do. At the same time, those categories of identity facilitate a sense of agency not by
accurately mirroring the social world, but by distorting it, abstracting away from the very facts of
complexity and interdependence that lead us to grasp for such categories in the first place. Recall
Antigone’s identifications, for example. Those identifications are secured as much by her
deliberate refusal of the vocabulary of the polis as by her invocation of the obligations of kinship.
For Antigone, to be a representative of the family, the divine law, and woman means not to be
compelled by the claims of the city and the human law, and not to be identified with masculinity;
and the converse is true for Creon. But it is precisely because identity emerges in this way out of
28

distinctions and oppositions that it remains tied to that from which it seeks to distinguish itself.
The practical possibilities and ways of being we disavow in the course of ordering ourselves into
the bearers of coherent “identities” do not, simply by virtue of that disavowal, cease to bear on
what we do: hence Antigone’s action takes on political significance at the very moment that she
attempts to fix its meaning firmly within the sphere of kinship. Antigone’s and Creon’s
blindness to parts of the meanings of their acts is no accident; it is a structural ignorance,
imposed by the nature of identification itself. It is the distinctive form of misrecognition
(Aristotle might have called it a hamartia) that afflicts the pursuit of recognition. 83
* * *
We are now in a position to return to the theme of tragic anagnôrisis. Aristotle first
defines the concept in chapter 11 of the Poetics, in the course of elaborating the nature and
components of the tragic plot: “A recognition is, as the very word implies, a change from
ignorance [agnoias] to knowledge [gnosin], and thus to either love or hate, in the personages
marked for good or evil fortune.” 84 Aristotle proceeds (both in this chapter and later) to list
some examples of recognition, including the recognition of Odysseus by his scar (1454b25–8);
Electra’s recognition of Orestes by “reasoning” in Aeschylus’s Choephori (1455a4–6); and
Oedipus’s recognition that he himself is his father’s killer (1452a32–3). It is tempting to
conclude on the basis of these examples that “recognition” in the Poetics has the same sense that
it did in the preceding account of the Antigone as a “struggle for recognition” (and hence the same
sense that it has in the contemporary normative discourse of recognition). After all, the examples
all seem to suggest that recognition is the recognition of an identity, either one’s own or
another’s. 85
29

This is not altogether wrong, but it is misleading. A somewhat different view emerges if
we situate the concept of recognition within Aristotle’s larger account of the structure of tragedy,
and tie it back to the account of êthos and action we have been developing. 86 First, it is
important to acknowledge that for Aristotle, recognition is closely linked to another component
of what he calls “complex” tragic plots, peripeteia or “reversal”; indeed, Aristotle says that the
“finest form” of recognition is that accompanied by reversal (1452a32–33). And reversal, in turn,
is a particular form of change (metabolê) in the fortune or circumstances of an actor; specifically,
it is a change that takes place through a relatively abrupt shift in the apparent trajectory of the
action. 87 The presence of such abrupt shifts or reversals in the action, in turn, can be understood
as Aristotle’s way of spelling out what it means for the events that make up a tragic plot to occur
“unexpectedly and at the same time in consequence of one another” (para tên doxan di’ allêla,
1452a4), which is a feature that he says characterizes the most effective tragic plots. 88 And this,
finally, bring us back to êthos and action, for it is precisely the gap between who we are and what
we do—what Halliwell calls the “disparity between the knowledge or intentions of the dramatic
characters and the underlying nature of their actions”—that drives this sort of plot, thick with
surprise but not merely accidental. Aristotle’s original claim about the order of priority of action
and êthos, and our parallel understanding of the impropriety of action with respect to identity,
thus turn out to be intimately connected to recognition: action’s impropriety is the one of the
central occasions for tragic anagnôrisis.
This way of situating recognition suggests a different account of its meaning. If the
priority of action over êthos and identity is the occasion for reversal and thus for recognition,
then perhaps we can understand recognition’s “shift from ignorance to knowledge” not only as
the discovery of someone’s true identity, but also and more fundamentally as what I would call
30

an “ontological” discovery, a shift from ignorance to knowledge about the real conditions of one’s
own existence and activity, and especially about the very relationship between êthos or identity
and action. On this account of tragic anagnôrisis, to be sure, recognition may indeed take place
through the discovery of something about one’s own or someone else’s “identity,” at least in the
ordinary sense that we have in mind when we say that Oedipus discovered the identity of his
father’s killer. But what differentiates this account of recognition from the contemporary
normative discourse of recognition (and from recognition as Antigone and Creon pursued it) is
that this sort of recognition does not satisfyingly consolidate and strengthen a practical identity,
or an identity in the thick sense we have been using here—that is, a coherent set of commitments
and values that enable an agent to know what to do. To the contrary, tragic recognition comes
after action, exposing the limits of practical identity, and shattering without yet reconstituting the
coherence of the identificatory scheme with which the agent has tried to govern his activity.
Tragic recognition, we might say, is the acknowledgment of finitude under the weight of a (failed)
effort to become sovereign through the recognition of identity.
This understanding of tragic anagnôrisis is borne out by the text of the Antigone, for
while the Antigone’s struggles for recognition do indeed fail, the play nevertheless contains
moments of “successful” (though unhappy) recognition in this second sense. The Antigone’s
tragic recognitions are not satisfying recognitions of authentic identities; instead, they are
mournful acknowledgments of human finitude, prompted by first-hand experience with the risks
and dangers that accompany the pursuit of satisfaction through recognition. We might at first
expect such moments of tragic recognition to consist in straightforward acknowledgments of guilt,
as for instance when Creon, frightened by Teiresias’s prophecies, accepts the chorus’s counsel
and rushes off to bury Polyneices and free Antigone. But if Creon’s reversal of his decision were
31

the paradigmatic recognition scene in the Antigone, we would look in vain for a corresponding
moment of recognition involving Antigone herself. The only parallel episode comes near the end
of Antigone’s self-lament, and is more equivocal than Creon’s plain reversal: “Should the gods
think that this [i.e., her punishment] is righteousness, in suffering I’ll see my error clear,” she
says. “But if it is the others who are wrong, I wish them no greater punishment than mine”
(926–29). To say that she confesses her guilt here would be to overstate the case, for she only
professes uncertainty about whether the gods will find her righteous or culpable. 89
Yet to suppose on these grounds that only Creon experiences a moment of recognition is
both to overlook something important about Antigone’s words and to misunderstand what this
moment of tragic recognition involves. In the first place, we must be careful to avoid a false
choice between an Antigone who straightforwardly confesses her guilt and an Antigone who
remains as resolute as ever to the end. In fact, both are untrue to the text. Both Antigone’s self-
lament and her admission of the possibility that the gods will judge her guilty represent real
transformations of Antigone’s earlier rigidity. In her opening speeches, Antigone had defiantly
welcomed the prospect of death (70–72), but now she mourns her fate, and in so doing, explicitly
acknowledges that her act has deprived her of the very goods she pursued, including philia and
motherhood (878–81, 918). And, in fact, by acknowledging that the question of the real
significance of her act is out of her hands and rests instead with the gods, Antigone expresses a
more sophisticated recognition of human finitude, of the gap between identity and action, than a
simple avowal of guilt would have indicated. After all, even after Creon changes his tune, he is no
more master of his fate than he had been earlier; in fact, his worst suffering is still to come.
Creon’s real moment of recognition comes after the deaths of his son and wife. As he laments his
fate and prays for a quick death, Creon’s own words reflect his belated awareness not of his
32

wrongdoing, exactly (for he has already seen that) but of the gap between his intentions and his
actions: “Take me away at once,” he pleads, “the frantic man who killed my son, against my
meaning” (1339–41). What Antigone and Creon have recognized, in different ways, is a version
of the predicament described by the Chorus in its famous closing words (1343–52): to avoid the
catastrophes that action’s improprieties bring, we would need to possess as actors a practical
wisdom that we acquire only in retrospect, and too late. 90
NOTES
I owe special thanks to Danielle Allen, Seyla Benhabib, Peter Euben, Jill Frank, Bonnie Honig,
and Patrick Riley for their generous and helpful comments on earlier drafts. This is still a work
in progress, and I have not yet been able to respond as fully as I intend to all the suggestions these
readers have made.
1. Hannah Arendt, The Human Condition (Chicago: University of Chicago Press, 1953), 193.
2. Jean-Jacques Rousseau, Politics and the Arts: Letter to M. D’Alembert on the Theatre, ed. and
trans. Allan Bloom (Ithaca: Cornell University Press, 1960), 126.
3. The passage is quoted by Charles Taylor, “The Politics of Recognition,” in Multiculturalism:
Examining the Politics of Recognition, ed. Amy Gutmann (Princeton: Princeton University Press,
1994), 47–48.
4. On recognition, see Aristotle Poet. 1452a29–b8; on its importance to “the finest form of
tragedy,” Poet. 1452b30–34. Except where otherwise noted, quoted passages from the Poetics in
this essay follow Ingram Bywater’s translation, in Aristotle, On the Art of Poetry (Oxford:
Clarendon, 1909). I do make two consistent and small changes in Bywater’s translation without
further notice: I reverse his practice of capitalizing the first letters of such words as “tragedy,”
“plot,” and “character”; and I change his idiosyncratic and less literal rendering of anagnôrisis as
“Discovery,” substituting “recognition.” In working with the Poetics I have also benefited from
the texts, translations, and/or commentaries contained in: Aristotle, Poetics, ed. D. W. Lucas
(Oxford: Clarendon Press, 1972); Stephen Halliwell, The Poetics of Aristotle: Translation and
Commentary (Chapel Hill: University of North Carolina Press, 1987); Gerald F. Else, Aristotle’s
Poetics: The Argument (Cambridge, MA: Harvard University Press, 1957); Aristotle, Poetics, ed.
and trans. Gerald F. Else (Ann Arbor: University of Michigan Press, 1970).
5. For this history, see Terence Cave, Recognitions: A Study in Poetics (Oxford: Clarendon Press,
1990).
33

6. This ambivalence is registered in Aristotle’s definition of anagnôrisis, which he says is a
“change from ignorance to knowledge, and thus to either love or hate, in the personages marked
for good or evil fortune” (Poet. 1452a29–32). It is also nicely reflected in Jocasta’s next-to-last
words to Oedipus, “May you never know who you are!” (Soph. Oed. Tyr. 1068). Our tendency
to regard recognition as an unequivocal good makes it impossible to grasp this exclamation as an
expression of concern and goodwill, as it seems to be.
7. For analogous uses of “ontological” see Stephen K. White, Sustaining Affirmation: The
Strengths of Weak Ontology in Political Theory (Princeton: Princeton University Press, 2000);
and the discussion of “ontopolitical interpretation” in Connolly, “Nothing is Fundamental,”
chap. 1 in The Ethos of Pluralization (Minneapolis: University of Minnesota Press, 1995).
8. In Aristotle’s terms, tragedy is an “imitation of an action”: Poet. 1450a16.
9. Cf. J. Peter Euben, The Tragedy of Political Theory: The Road Not Taken (Princeton:
Princeton University Press, 1990), 56.
10. Both Axel Honneth (The Fragmented World of the Social [Albany: SUNY Press, 1995], 250)
and Taylor (“The Politics of Recognition,” 26) say that misrecognition can “cripple” its victims.
For the unmediated connection between identity and agency, see Avishai Margalit and Moshe
Halbertal, “Liberalism and the Right to Culture,” Social Research 61, no. 3 (Fall 1994): 491–510;
498); for the mediated connection, see Will Kymlicka, Multicultural Citizenship: A Liberal Theory
of Minority Rights (Oxford: Clarendon Press, 1995), chap. 5 (though I believe Kymlicka is pushed
further toward Margalit’s and Halbertal’s position than he acknowledges in his effort to justify
not only a right to culture in general but a right to “one’s own” culture: since what the latter idea
adds is the notion of cultures as plural and to at least some degree mutually exclusive, it demands
a shift in focus away from the choices culture contains (which it might well share with others)
and toward the options it excludes (which are what marks it off as unique).
11. Michael J. Sandel, Liberalism and the Limits of Justice (Cambridge: Cambridge University
Press, 1982).
12. Of course, not only the Poetics is implicated here: the Nicomachean Ethics (especially the
discussion of virtue, misfortune, and eudaimonia in book 1, ch. 10); the Metaphysics (esp. book
9); and the Politics are also all important, as are (for me) Arendt’s uses of some of these
Aristotelian texts in her account of action in The Human Condition, I intend to expand the
discussion of these other texts in a future version of this paper.
13. Parenthetical citations to the Antigone refer to line numbers, and unless otherwise noted,
quotations from the play follow Elizabeth Wyckoff’s translation, in the first edition (only) of
David Grene and Richard Lattimore, eds., The Complete Greek Tragedies, vol. II, Sophocles
(Chicago: University of Chicago Press, 1959).
34

14. Polyneices’ grievance is discussed in Aesch. Septem 631–52 and Soph. Oed. Col.
1284–1307. On the death of the brothers see Aesch. Septem 808–14.
15. In Aeschylus’s version of the story, Eteocles and Polyneices died childless; see G. O.
Hutchinson’s discussion in his note to line 903 in Aeschylus, Septem contra Thebas (Oxford:
Clarendon Press, 1985). In other accounts, Polyneices left a son, Thersander, who, together with
the other sons of the Seven (the “Epigoni”) later successfully conquered Thebes. For a summary
of sources see Timothy Gantz, Early Greek Myth: A Guide to Literary and Artistic Sources
(Baltimore: The Johns Hopkins University Press, 1993), 522–25.
16. Terence Cave has even suggested that the burst of interest in the Antigone that began in the
late eighteenth century, and which found its most influential expression in Hegel, helped bring
about “the eclipse of anagnorisis as a central tragic concept” in modern poetics. Recognitions,
154.
17. Indeed, the Antigone is not only full of claims to recognition; it is also surrounded by them.
Readers of the play commonly make claims to recognize Antigone and Creon—to know who
they are and what they stand for. In a classic essay, Jean Bethke Elshtain argues that Antigone is
a representative of “primordial family morality”—an identity that, in the form of “maternal
thinking,” is still available to be taken up in acts of resistance to modern Creons (“Antigone’s
Daughters,” democracy 2, no. 2 [April 1982]: 53, 58–59). In a response to Elshtain, Mary Dietz
claims instead that Antigone represents “the customs and traditions of a collective civil life, an
entire political ethos” (“Citizenship with a Feminist Face: The Problem with Maternal
Thinking,” Political Theory 13, no. 1 [February 1985]: 29). Rather than taking sides in this
debate about what Antigone, properly understood, represents, I suggest that we try to account
for the fact that Elshtain’s and Dietz’s seemingly incompatible accounts of Antigone’s identity
are both partly persuasive. We can do so by focusing less exclusively on who Antigone is and
attending instead to how who she is relates to what she does. This angle can help us understand
how Antigone’s deeds do not smoothly flow from and reinforce, but exceed and disturb our (and
her) conception of her identity.
18. See Larry J. Bennett and William Blake Tyrrell, “Sophocles’ Antigone and Funeral Oratory,”
American Journal of Philology 111, no. 4 (Winter 1990): 442. On the question of whether this
topos preceded Sophocles, see Christiane Sourvinou-Inwood, “Assumptions and the Creation of
Meaning: Reading Sophocles’ Antigone,” Journal of Hellenic Studies 109 (1989): 134–48, arguing
that Sophocles introduced the idea that “the total denial of burial offends against the gods” (143),
but also the response by Helene Foley, “Tragedy and Democratic Ideology: The Case of
Sophocles’ Antigone,” in History, Tragedy, Theory: Dialogues on Athenian Drama, ed. Barbara
Goff (Austin: University of Texas Press, 1995), arguing that the topos of the recovery of the
bodies “could well have been in place before the Antigone was composed” (140). For a summary
of versions of the story and textual evidence concerning norms of burial, see appendix C to D. A.
Hester, “Sophocles the Unphilosophical: A Study in the Antigone,” Mnemosyne, 4th ser., 24
(1971): 54–55.
35

19. See for instance Lysias 2.7–10 and the discussion in Bennett and Tyrrell, “Sophocles’
Antigone and Funeral Oratory,” 443.
20. The focus on the brothers may not be wholly original with Sophocles, since the ending of
Aeschylus’s Seven against Thebes also describes an edict concerning Polyneices and Eteocles,
and depicts Antigone’s resolve to bury Polyneices despite the edict (1010–78). But the
authenticity of the ending is contested, and some argue it to have been a late addition based upon,
not preceding, the Antigone; for a discussion of the issue see Hutchinson’s edition of Aeschylus,
Septem contra Thebas, 209–11.
21. See W. K. Lacey, The Family in Classical Greece (Auckland: University of Auckland, 1980),
148; on the legal obligation see S. C. Humphries, The Family, Women, and Death: Comaprative
Studies, 2nd ed. (Ann Arbor: University of Michigan Press, 1993), 83.
22. The text of the Antigone does nothing to contradict Creon’s view of Polyneices as traitor and
Eteocles as patriot; even the chorus describes Polyneices as pursuing a “dubious quarrel” (110).
23. Thucydides, for instance, notes that Athenian law prohibited the burial of traitors in Attic
soil (1.10); see the discussion of this law in Sourvinou-Inwood, “Assumptions and the Creation
of Meaning,” 137–38, esp. n. 20; and Hester, “Sophocles the Unphilosophical,” 55.
24. Donna C. Kurtz and John Boardman, Greek Burial Customs (London: Thames and Hudson,
1971), 260; see also Christiane Sourvinou-Inwood, ‘Reading’ Greek Death: To the End of the
Classical Period (Oxford: Clarendon Press, 1995), 177–79; Gail Holst-Warhaft, Dangerous
Voices: Women’s Laments and Greek Literature (London: Routledge, 1992), chap. 4; for a
modern parallel see Arendt, “On Humanity in Dark Times: Thoughts about Lessing,” in Men in
Dark Times (New York: Harvest/HBJ, 1968), 20–22. Hegel’s discussion of the work of
immortalization performed by burial in his discussion of Greek ethical life in The Phenomenology
of Spirit, trans. A. V. Miller (Oxford: Oxford University Press, 1977), 270–71, is also useful here.
25. The locus classicus is Sigmund Freud, “Mourning and Melancholia,” in Collected Papers,
vol. 4 (New York: Basic Books, 1959). See also Eric L. Santner, Stranded Objects: Mourning,
Memory and Film in Postwar Germany (Ithaca: Cornell University Press, 1990), chap. 1; Peter
M. Sacks, The English Elegy: Studies in the Genre from Spenser to Yeats (Baltimore: The Johns
Hopkins University Press, 1985), chap. 1. Arendt borrows a line from Isak Dinesen that
captures the importance of lament to the mourner: “All sorrows can be borne if you put them
into a story or tell the story about them.” Arendt used these words as the epigraph to her
chapter on “Action” in The Human Condition (Chicago: University of Chicago Press, 1953), but
she also quotes them in her essay “Isak Dinesen, 1885–1963,” and adds her own gloss: “The
story reveals the meaning of what otherwise would remain an unbearable sequence of sheer
happenings” (Men in Dark Times, 104). Notably, the final sentence of Arendt’s essay on
Dinesen reads like a loose translation of the chorus’s closing words in the Antigone: “Wisdom is a
virtue of old age, and it seems to come only to those who, when young, were neither wise nor
prudent” (109).
36

26. See here Sacks’s discussion of consolation and substitution in The English Elegy, 6–7.
27. This aspect of funeral ritual is reflected in Hegel’s gloss on the practice of burial in Greece as
a way of humanizing death by supplementing a natural and “irrational” event with “an action
consciously done” G. W. F. Hegel, Phenomenology of Spirit, trans. A. V. Miller (Oxford: Oxford
University Press, 1977), 270.
28. The significance of philos and ekhthros in the Antigone has been extensively discussed. I
have benefited most from Simon Goldhill, Reading Greek Tragedy (Cambridge: Cambridge
University Press, 1986), 79–106, to whom I owe in particular the insight into the connection
between philia and recognition (85). See also Mary Whitlock Blundell, Helping Friends and
Harming Enemies: A Study in Sophocles and Greek Ethics (Cambridge: Cambridge University
Press, 1989), 106–48; David Konstan, “Greek Friendship,” American Journal of Philology 117
(1996): 82–85; Charles Segal, Tragedy and Civilization: An Interpretation of Sophocles
(Cambridge, MA: Harvard University Press, 1981), 185–86; and Matt Neuburg, “How Like a
Woman: Antigone’s ‘Inconsistency,’” Classical Quarterly 40, no. 1 (1990): 70–76.
29. See the sources in the preceding note as well as the relevant entries in Henry George Liddell
and Robert Scott, A Greek-English Lexicon, 9th ed., rev. Henry Stuart Jones and Roderick
McKenzie (Oxford: Oxford University Press, 1968).
30. See here especially Goldhill, Reading Greek Tragedy, 79–83. The connection I draw here
between philia and recognition should shed some light on the long-running debate over whether
the use of philos indicates subjective “affection” or an objective, “ascribed status”—which I
believe to be a false choice. David Konstan, for instance, has recently argued against Goldhill and
others that when Antigone calls Polyneices philos, she does not thereby “indicat[e] kinship as
the grounds for [her] behavior,” but merely expresses her affection for him. Her use of philos
could not possibly refer to the ascribed status of kinship, Konstan argues, because this would
make “nonsense of Antigone’s description of Ismene as an enemy.” Konstan admits, however,
that Antigone’s harshness toward her sister may be intended by Sophocles to express “the irony
by which a sister, normally dear, is rendered inimical”—and the emphasized phrase contains the
key to the problem. Contra Konstan, this irony only works if Antigone’s use of philos not only
indicates affection but also implies some view about why that affection is owed—for instance, the
view that kin are “normally dear.” This need not mean, in Konstan’s terms, that philos simply
“signifies” kin—the way philos works cannot be understood from within a simply designative
theory of language. Rather, the use of philos—like “recognition” itself—is both epistemic and
normative; it both makes reference to objective relationships and announces a feeling of affection
grounded in those relationships. See Konstan, “Greek Friendship,” 83–84. [read Belfiore,
Murder among Friends, ch. 1]
31. This relationship between the use of philia and the recognition of identity is corollary to the
relationship Aristotle describes in his definition of anagnôrisis in the Poetics. There, Aristotle
says that recognition is, “as the very word implies, a change from ignorance to knowledge, and
37

thus either to love [philian] or hate [ekhthran] in the personages marked for good or evil fortune.”
Poet. 1452a29–32. In the Antigone, saying who one is takes the form of locating oneself in a
certain network of relationships of philia and ekhthra; likewise, in the Poetics, discovering who
one is has the effect of transforming one’s understanding of who is philos and who
ekhthros—although, as I will suggest in the last section of this essay, tragic anagnorisis is not
simply the recognition of one’s identity, at least as the contemporary politics of recognition
encourages us to understand that phrase.
32. On the interpretation of this line see R. C. Jebb’s note in his edition of Sophocles, The
Antigone, part 3 of The Plays and Fragments, 3rd ed. (Cambridge: Cambridge University Press,
1900), 10–11, as well as Blundell, Helping Friends and Harming Enemies, 107 n. 5.
33. On the distinction between the chthonian and Olympian in the context of the Antigone, see
Segal, Tragedy and Civilization, 171.
34. Ibid., 183–84. When contested, citizenship was established by demonstration of enrollment
in the phratry, a patrilineal kinship organization into which daughters were not introduced.
35. Ibid., 184.
36. See Foley, “Tragedy and Democratic Ideology,” 146 n. 11 (responding to Sourvinou-
Inwood, “Assumptions and the Creation of Meaning,” 140). On the roles of men and women at
funerals see Robert Garland, The Greek Way of Death (Ithaca: Cornell University Press, 1985),
chap. 3; Humphries, The Family, Women and Death, 83–88.
37. See Foley, “Tragedy and Democratic Ideology,” 146 n. 11 (responding to Sourvinou-
Inwood, “Assumptions and the Creation of Meaning,” 140). On the roles of men and women at
funerals see Robert Garland, The Greek Way of Death (Ithaca: Cornell University Press, 1985),
chap. 3; Humphries, The Family, Women and Death, 83–88.
38. Jebb’s and Grene’s translation; Wyckoff translates philos kakos as “a friend no friend.”
39. See his statements to the chorus (484–85), to Antigone (525), and to Haemon (678–80).
40. This view is often associated with Hegel, and has provoked critics of Hegel to complain that
by reducing Creon and especially Antigone too completely to personifications of opposed social
forces, he overlooks the ways in which, for example, Antigone “breaks out of her assigned
sphere” (Mills, Woman, Nature and Psyche, 35). In fact, as I argue in a detailed reading of Hegel
elsewhere, these criticisms ignore a sharp and meaningful break (which occurs between
paragraphs 463 and 464) in Hegel’s discussion of the Antigone in the Phenomenology. In the
passages up to and including par. 463, Hegel describes the background structure of ancient Greek
Sittlichkeit, laying out the various ethical powers as though they formed a “whole [which] is a
stable equilibrium of all the parts” (par. 462). Characteristically, however, Hegel’s point is
precisely to reveal the limitations of what initially appears to be a satisfactory view or a coherent
38

practice; so beginning in par. 464, he moves from offering a general account of the structure of
Greek ethical life to personifying its constituent powers: here, Antigone and Creon, as individual
agents who commit “deeds,” come on the scene, and in so doing, they “disturb the peaceful
organization and movement of the ethical world”—that is, they reveal, precisely by their
improprieties, the contradictions and limitations that afflict Greek ethical life. So while Hegel
does indeed assign Antigone and Creon to different spheres, he also highlights the ways in which
they do break out of the spheres to which they are assigned; indeed, his reading depends on it.
41. Martha C. Nussbaum, The Fragility of Goodness: Luck and Ethics in Greek Tragedy and
Philosophy (Cambridge: Cambridge University Press, 1986), 63–64.
42. On Antigone’s treatment of Ismene see Blundell, Helping Friends and Harming Enemies,
111–15.
43. On the gendered architecture of the oikos, see Eva C. Keuls, The Reign of the Phallus: Sexual
Politics in Ancient Athens (New York: Harper and Row, 1985), 93–97.
44. Emily Vermeule notes that in Sophocles’ description the cave has all the architectural
features of a Bronze age chamber tomb, which was itself called an “oikos for the dead”; see
Aspects of Death in Early Greek Art and Poetry (Berkeley: University of California Press, 1979),
54. On the tomb as bedroom or “bridal-chamber” in the Antigone, see Rush Rehm, Marriage to
Death: The Conflation of Wedding and Funeral Rituals in Greek Tragedy (Princeton: Princeton
University Press, 1994), chap. 4, and Richard Seaford, “The Imprisonment of Women in Greek
Tragedy,” Journal of Hellenic Studies 110 (1990): 76–90.
45. Nicole Loraux shows a distinction between men’s epitaphs, which acknowledged and
recorded distinctiveness, with women’s epitaphs, which praised women (if at all) by assimilating
them to a generic type; the “orthodox form” of a woman’s epitaph, Loraux says, is indicated by
the example: “Supposing that feminine virtue still exists in the human race, she partook of it.”
Tragic Ways of Killing a Woman, trans. Anthony Forster (Cambridge, MA: Harvard University
Press, 1987), 27. On the avoidance and suppression of living women’s names, see John Gould,
“Law, Custom and Myth: Aspects of the Social Position of Women in Classical Athens,”
Journal of Hellenic Studies 100 (1980): 45.
46. Neuburg, “How Like a Woman,” 66–67. See also Sheila Murnaghan, “Antigone 904–20 and
the Institution of Marriage,” American Journal of Philology 107, no. 2 (Summer 1986): 192–207.
47. As Christina Elliot Sorum puts it, “in choosing to perpetuate her alliance with her origin,
Antigone rejects her own potential as an origin.” See “The Family in Sophocles’ Antigone and
Electra,” Classical World 75, no. 4 (March–April 1982): 206. Antigone’s grouping of children
with husbands as among the “replaceable” kin (909–10) might seem to undermine my association
of children with blood-ties, since the distinction between blood-ties and marriage-ties is taken by
both Neuburg (73) and Murnaghan (198) to correspond to the distinction between irreplaceable
and replaceable kin. The point, however, is that children represent the point of intersection
39

etween (and therefore the inseparability of) the two kinds of tie: even Neuburg refers to
Haemon, for instance, as bound to Creon by blood (74), even though according to Antigone’s
distinctions, Haemon (as a child) should be replaceable. The problem here is that “blood-tie vs.
marriage-tie” does not correspond perfectly to “irreplaceable vs. replaceable,” because blood and
marriage ties are interrelated in a way that the terms of the latter distinction, which really is a
simple opposition, are not. But even if lines 904–20 involve the issue of replaceability more than
an opposition between ties of blood and marriage, this does not render the lines nonsensical:
given the function of mourning in restoring agency in the face of loss, the fact of the
irreplaceability of a loved one may make the imperative of mourning unconditional, since the
investment in the lost person cannot as easily be transferred to a surrogate object. As
Murnaghan cautions, though, “we should not search for an interpretation that would eliminate
our dissatisfaction with Antigone’s words” (207); they are the words of a woman who is herself
dissatisfied, lamenting the losses imposed upon her by an impossible choice that no
rationalization can clarify without remainder.
48. On this as an expression of specifically filial devotion, see Nussbaum, The Fragility of
Goodness, 62; Joan V. O’Brien, Guide to Sophocles’ Antigone (Carbondale: Southern Illinois
University Press, 1978), 77.
49. On the military language, see Jebb’s note to line 640 and O’Brien, Guide to Sophocles’
Antigone, 79, from which the quoted words are taken.
50. Segal, “Lament and Closure in Antigone,” 131.
51. As many commentators have observed, the actual order of Creon’s actions reverses the order
in which the Chorus recommends he proceed (1100–1101); for a discussion of this “blemish” see
Jebb’s introduction to his edition of The Antigone, xviii–xx.
52. The reference of these lines to the bodies of the sons of “the cities that you fought in war” as
a source of pollution forgivably conflates Polyneices’ corpse with the other Argive dead.
53. In fact, Sophocles has already introduced the theme of the difference between the actor and
the significance of the act in another episode involving a messenger: the guard who first informs
Creon that Polyneices’ corpse has been buried is initially suspected of having committed the
crime himself, though he protests that he “never did the deed” (321).
54. Arendt, The Human Condition, 184. Arendt specifically mentions the use of messengers
within tragedy as an illustration of the gap between perspective of the actor and the perspective
of the narrator in “The Concept of History,” in Between Past and Future (New York: Penguin,
1968), 45.
55. On the translation of the Poetics used here see note 3 above. For a good discussion of
Aristotle’s vocabulary, see Stephen Halliwell, Aristotle’s Poetics (Chapel Hill: University of
North Carolina Press, 1986), 138–42. Aristotle sometimes uses praxis or praxeis to refer to the
40

action or actions of persons in an ordinary sense; at other times, he uses praxis in the singular to
refer to the whole course of action that is imitated in the plot (muthos) of a drama; and the
muthos, in turn, is said to be a combination or arrangement of pragmata—itself cognate with
praxis, and which is usually translated as “incidents” or “events.”
56. This claim has sometimes been a source of consternation because it runs counter to our post-
Romantic preoccupation with what Hegel called the “subjective inner life” of dramatic characters
(Aesthetics: Lectures on Fine Art, trans. T. M. Knox, vol. 2 [Oxford: Clarendon Press, 1975],
1223). As one of Aristotle’s editors writes:
A[ristotle]’s preference for plot as opposed to character has won little approval
from most of the critics from the late nineteenth century onwards. To the
generations which were profoundly influenced by Bradley’s Shakespearian
Studies it was common doctrine that, as Granville Barker once put it, the purpose
of drama was to portray character. Interest in the inner life of the individual,
which had been developed by the great novelists of late Victorian times in
England, France, and Russia, caused exaggerated attention to traits of personality
which could be perceived in Shakespeare and contributed to the spread of the
belief that they must be contained, could one but find them, in all great drama.
D. W. Lucas, “Commentary” to Aristotle, Poetics (Oxford: Clarendon Press, 1972), 109–110.
For a critique of anachronistic tendencies (such as the focus on the “tragic hero”) on the part of
modern readers of the Poetics and of Greek tragedy, see John Jones, On Aristotle and Greek
Tragedy (London: Chatto and Windus, 1962), chap. 1.
57. For one interesting example of such a reading see Elizabeth Belfiore’s treatment of the
relationship between the Poetics and modern narrative theory: “Narratological Plots and
Aristotle’s Mythos,” Arethusa 33, no. 1 (Winter 2000): 37–70. In an earlier work, however,
Belfiore argues that Aristotle’s prioritization of action in the Poetics is evidence of a split
between his dramatic theory and his understanding of “real life,” where êthos does indeed have
priority over action; I discuss and criticize this claim in more detail below.
58. On controversies surrounding the text of this passage see Nussbaum, The Fragility of
Goodness, 378–79 and accompanying notes.
59. They echo, for example, the discussion of eudaimonia as an activity in NE 1.7, 1098a15–20,
as has been noted by Amelie Oksenberg Rorty, “The Psychology of Aristotelian Tragedy,” in
Essays on Aristotle’s Poetics (Princeton: Princeton University Press, 1992), 20 n. 12.
60. Nussbaum, The Fragility of Goodness, 379–82; cf. Halliwell, Aristotle’s Poetics, 157.
Another way of putting this point is to say that tragedy is the “imitation of an action” not only
in the obvious sense that it represents some sequence of events on the stage, but also because its
very structure mirrors something important about the nature of action itself.
41

61. This use of Aristotle’s Poetics might be thought to follow through on Nussbaum’s suggestion
that insights drawn from the discussion of action and character in book 6 could provide the basis
for a reinterpretation of the concept of anagnôrisis: see The Fragility of Goodness, 382.
62. Arendt, The Human Condition, 244. The following account is heavily informed by Arendt’s
account of action (especially 188–92, 236–47). For a good synoptic discussion of Arendt’s
concept of action, see Maurizio Passerin d’Entrèves, The Political Philosophy of Hannah Arendt
(New York: Routledge, 1994), 64–100.
63. This might be taken to be the implication of Elizabeth Belfiore’s claim that “in the Poetics,
êthos as a part of tragedy is always an indication of choice”—i.e., of prohairesis—“and never
includes a broader set of ‘characteristic peculiarities,’ as êthos sometimes does in the Rhetoric”
(Tragic Pleasures: Aristotle on Plot and Emotion [Princeton: Princeton University Press, 1992],
97). Blundell has suggested that even in the Rhetoric, such attributes as age, sex, and nationality
are not to be understood as aspects of êthos but as influences upon and potenial indicators of it:
“Êthos and Dianoia Reconsidered,” in Amelie Oksenberg Rorty, ed., Essays on Aristotle’s
Poetics (Princeton: Princeton University Press, 1992), 164.
64. Charles Chamberlain, “From ‘Haunts’ to ‘Character’: The Meaning of Êthos and its Relation
to Ethics,” Helios 11, no. 2 (Winter 1984): 97–108.
65. Ibid., 101; the passage from Thucydides (2.61.4) is cited at 106 n. 44; see Thucydides,
History of the Peloponnesian War, ed. M. I. Finley, trans. Rex Warner (London: Penguin, 1972),
160.
66. Chamberlain also notes that “êthos comes from the Indo-European root *swedh, see also in
the Latin suus and suesco, meaning ‘one’s own’ or ‘proper’” (104 n. 8). Exploiting this
connection, we might say that something’s êthos is just that which leads it to do what is proper
to it, as the sort of thing it is.
67. I am bracketing controversies over the complicated relationship of êthos to thought or
intellect (dianoia) both in the Poetics and in Aristotelian ethics, since these aren’t relevant to the
aspects of êthos I’m emphasizing here. For discussion and bibliography, see Mary Whitlock
Blundell, “Êthos and Dianoia Reconsidered,” in Amelie Oksenberg Rorty, ed., Essays on
Aristotle’s Poetics (Princeton: Princeton University Press, 1992).
68. Nancy Sherman, The Fabric of Character: Aristotle’s Theory of Virtue (Oxford: Clarendon
Press, 1989), 1.
69. The Poetics itself says that êthos (here used in the at-one-remove sense of
“characterization,” or the dramatic depiction of “character”—on this see Halliwell, Aristotle’s
Poetics, 150–51) is that which “shows us the nature of a prohairesis”; certainly Antigone’s and
Creon’s efforts to ground their actions in their identities show us aspects of the nature of their
choices. (Poet. 1450b8–9; I have used Halliwell’s translation because Bywater unfortunately
42

translates prohairesis as “moral purpose.”) Indeed, in the speech quoted above in which Haemon
exhorts his father not to “have one mind, and one alone/that only your opinion can be right”
(705), “mind” renders “êthos.”
70. Sherman, The Fabric of Character, 1.
71. Alasdair MacIntyre, After Virtue, 2nd ed. (Notre Dame: University of Notre Dame Press,
1984), 149, 148.
72. Elizabeth S. Belfiore, Tragic Pleasures: Aristotle on Plot and Emotion (Princeton: Princeton
University Press, 1992). In “real-life situations,” Belfiore says, “action is caused by êthos and
thought”; in drama, on the other hand, where the action is devised by the poet and the êthos is
added later to flesh out the meaning of the action, the order of priority is reversed (89). On this
basis, Belfiore criticizes Nussbaum’s claim that for Aristotle “the great tragic plots explore the
gap between our goodness and our good living,” claiming that this represents a “charactercentered
view of tragedy opposed to Aristotle’s plot-centered view” (91). But, as I shall suggest,
Nussbaum’s view and ones like it are hardly character-centered in the relevant sense of granting
priority to character; quite to the contrary, Nussbaum focuses on exactly those aspects of
tragedy that demonstrate the limits of character’s capacity to determine action.
73. I owe my understanding of this point to conversations with Jill Frank; for an excellent and
concise discussion of the relationships among dunamis, energeia, hexis, and praxis, see her
“Democracy and Distribution: Aristotle on Just Desert,” Political Theory 26, no. 6 (December
1998), esp. 795–96.
74. Nussbaum, The Fragility of Goodness, 382; on vulnerability, see 380–81, as well as
Halliwell, Aristotle’s Poetics, chap. 7.
75. For some other discussions of the first question in the context of Aristotle, see Nussbaum,
The Fragility of Goodness, esp. chaps. 11–12; Rorty, “The Psychology of Aristotelian Tragedy;
Aryeh Kosman, “Acting: Drama as the Mimêsis of Praxis,” in Rorty, ed., Essays on Aristotle’s
Poetics; Sherman, “Hamartia and Virtue,” in Rorty, ed., Essays on Aristotle’s Poetics; and
Halliwell, Aristotle’s Poetics, chap. 7.
76. Though this does not mean that will and agency are only possible sources of injury and
damage because of the unpredictability of action. Often enough, of course, injury and damage are
the result of successful, deliberate projects.
77. On this see Thomas Nagel’s discussion (in the context of an argument about luck and
responsibility in ethics) of the apparently irresolvable split between the “internal” and “external”
perspectives on our actions, in “Moral Luck,” in Mortal Questions (Cambridge: Cambridge
University Press, 1979), esp. 37; as well as Kosman, “Acting,” 65. On how concepts of
responsibility can be regarded as socially constructed but nevertheless effective and non-
43

arbitrary, see Marion Smiley, Moral Responsibility and the Boundaries of Community (Chicago:
University of Chicago Press, 1992).
78. This is not to say that natural disasters are always properly understood as the entirely
arbitrary action of wholly external forces upon blameless human beings: what we experience as
the sheer contingency of nature is sometimes closely related to our own activity and modes of
interaction with nonhuman forces—as, for example, when homes built on “former” floodplains
are destroyed by floods; or when the overprescription of highly successful antibiotics contributes
to the development and spread of drug-resistant strains of bacteria. For more detailed examples
of such tragic interactions between human beings and nature, especially in modern scientific
forestry and agriculture, see James C. Scott, Seeing Like a State: How Certain Schemes to
Improve the Human Condition Have Failed (New Haven: Yale University Press, 1998).
79. On the inferiority of tragic plots based on “mere chance,” see Arist. Poet. 9, 1452a1–11;
good discussions can be found in Sherman, “Hamartia and Virtue”; Rorty, “Psychology of
Aristotelian Tragedy”; Bernard Yack, The Problems of a Political Animal: Community, Justice,
and Conflict in Aristotelian Political Thought (Berkeley: University of California Press, 1993),
chap. 8.
80. On this implication of freedom, see Arendt, The Human Condition, 244, and “What is
Freedom,” in Between Past and Future, 169–71. Amelie Rorty mentions that “reversals” of
action are “most likely to occur in the interaction among several characters, each acting from his
or her own intentions”: “The Psychology of Aristotelian Tragedy,” 11.
81. Seyla Benhabib, also referring to Arendt, calls this the “interpretive indeterminacy” of
action. See her Critique, Norm, and Utopia (New York: Columbia University Press, 1986), 136.
82. Although causal contingency also plays a role in the tragedy, I think the account of
impropriety and identity I give below can ultimately be described as a specific modality or
manifestation of the way in which human plurality deprives us of sovereignty over the
significance of our actions.
83. Aristotle’s use of hamartia occurs in Poet. 13, 1453a10; for a good discussion see Sherman,
“Hamartia and Virtue”; and for an account parallel to mine of hamartia as a kind of blindness or
ignorance that arises precisely from an agent’s own virtues of character, see Rorty, “Psychology
of Aristotelian Tragedy,” 11. Rorty’s account seems distinctive to me in suggesting that the
vulnerabilities of character may arise precisely from the best moments of character itself; other
accounts of character’s vulnerability, such as Nussbaum’s, focus on the ways in which
circumstances external to character itself may render character insufficient for eudaimonia; for
example, Nussbaum (334–35, 380–81) focuses on the cases like Priam’s (in which misfortune
deprives someone of the resources or objects necessary for good activity); Oedipus’s (in which
the “luck of circumstances” makes it the case that the description under which someone intended
his act was, through no fault of his own, not the “morally salient description of what took
place”); and Agamemnon’s (in which circumstances require someone to choose, in full knowledge
44

of all the dimensions of his act, to do something that is in one sense good and in another sense
terrible). Rorty’s account, and the cases of Antigone and Creon, seem to suggest another
possibility, in which it is precisely in becoming and being a person of a certain êthos, or in
ordering one’s activity in accord with a certain identity, that someone becomes, perhaps
culpably, ignorant of the other potentially salient descriptions of his action.
84. Arist. Poet. 11, 1452a29–32.
85. Here I dissent from John Jones’s effort to correct the overemphasis on character in the
interpretation of tragedy by claiming that “the text makes it plain that we can’t” read anagnôrisis
as the recognition of an individual’s identity. In my view, it’s too strong to say that “the centre
of gravity of Aristotle’s terms is situational and not personal,” for Aristotle is concerned with
both, or more precisely with the relationship between them: we might say that he is concerned
with what becomes of persons under the weight of certain distinctive types of situation. (On
Aristotle and Greek Tragedy, 15–16). As the examples indicate, there is some sense in which it’s
impossible not to say that recognition is the recognition of a person (on this see Else, Aristotle’s
Poetics: The Argument, 352–53); the point will be that this is not the recognition of a coherent
practical identity in the sense (and with the unambiguously positive valence) presupposed by the
politics of recognition.
86. Textually, the following account works roughly backward from the definition of anagnôrisis
in chap. 11 to the immediately preceding definition of peripeteia, also in ch. 11; and finally to the
account of plots that occur “unexpectedly and at the same time in consequence of one another” in
chap. 9.
87. On the suddenness of peripeteia, see Else, Aristotle’s Poetics: The Argument, 345; Else
persuasively interprets peripeteia as a subset of metabolê at 343.
88. On the importance of this phrase and its connection to peripeteia and anagnôrisis, see Else,
Aristotle’s Poetics: The Argument, 329ff; Halliwell, Aristotle’s Poetics, 212.
89. On this point see Patricia Jagentowicz Mills, Woman, Nature and Psyche (New Haven: Yale
University Press, 1987), 28–29; Mills seems to me, however, to be trapped by the false choice I
identify in the next paragraph.
90. For an interesting reading of the Antigone in terms of belatedness, see Ulf Heuner,
“Sophokles’ Antigone—zur tragischen Ironie von Zeit und Handlung,” Poetica 29, nos. 1–2
(1997): 1–25.
45