Schriften des Historischen Kollegs
Kolloquien 49
Demokratie, Recht und soziale Kontrolle im klassischen Athen
R. Oldenbourg Verlag München 2002
Demokratie, Recht und soziale Kontrolle im klassischen Athen
Herausgegeben von David Cohen unter Mitarbeit von Elisabeth Müller-Luckner
R. Oldenbourg Verlag München 2002
S c h r i f t e n des H i s t o r i s c h e n K o ll eg s herausgegeben von
L othar Gail in V erbindung mit
A rnold Esch, Etienne Frangois, Johannes F ried, Klaus H ildebrand, M anfred H ilderm eier, Jochen M artin, H einrich Nöth, U rsula Peters, W olfgang Q uin t und D ietm ar W illow eit
G eschäftsführung: G eorg K alm er R edaktion: E lisabeth M üller-L uckner
Das H istorische Kolleg fördert im Bereich der historisch orientierten W issenschaften G e
lehrte, die sich durch herausragende L eistungen in Forschung und Lehre ausgew iesen haben.
Es vergibt zu diesem Z w eck jährlich bis zu drei F orschungsstipendien und ein F örderstipen
dium sow ie alle drei Jah re den „Preis des H istorischen K ollegs“ .
Die Forschungsstipendien, deren V erleihung zugleich eine A uszeichnung für die bisherigen Leistungen darstellt, sollen den berufenen W issenschaftlern w ährend eines K ollegjahrcs die M ö glichkeit bieten, frei von anderen V erpflichtungen eine größere A rbeit abzuschließen.
Professor Dr. David C ohen (Berkeley, C al.) w ar - zusam m en mit Professor Dr. G erhard Be- sier (H eidelberg), Prof. Dr. W olfgang R einhard (F reiburg i.Br.) und Dr. Lutz K linkham m er (Rom ) - Stipendiat des H istorischen Kollegs im K ollegjahr 1997/98. Den O bliegenheiten der Stipendiaten gem äß hat D avid C ohen aus seinem A rbeitsbereich ein K olloquium zum Them a „D em ocracy, Law, and Social C on tro l in C lassical A thens“ vom 16. bis 19. Ju n i 1998 im H istorischen Kolleg gehalten. D ie Ergebnisse des K olloquium s w erden in diesem Band veröffentlicht.
Das Flistorische Kolleg, bisher vom Stiftungsfonds Deutsche Bank zu r Förderung der W is
senschaft in Forschung und Lehre und vom Stifterverband für die D eutsche W issenschaft ge
tragen, w ird ab dem K ollcgjahr 2000/2001 in seiner G rundausstattung vom Freistaat B ayern finanziert; seine Stipendien w erden aus M itteln des D aim lerC h rysler Fonds, der Fritz. T h y s
sen Stiftung, des Stifterverbandes und eines ihm verbundenen Förderunternchm cns dotiert.
Träger des Kollegs ist nunm ehr die „Stiftung zur F örderung der H istorischen Kom m ission bei der Bayerischen A kadem ie der W issenschaften und des H istorischen K ollegs“.
Die D eutsche B ibliothek - C IP E inheitsaufnahm c
D e m o k r a t i e , R e c h t u n d so ziale K o n t r o l l e im kla ss isc he n A t h e n hrsg. von David C ohen unter M itarb eit von E lisabeth M ü llcr-L uckner. - M ünchen : O ldenbou rg 2002
(Schriften des H istorischen Kollegs : K olloquien ; 49) ISBN 3-486-56662-8
© 2002 O ldenbourg W issenschaftsverlag G m bH , M ünchen Rosenheim er Straße 145,D -81671 M ünchen
Internet: http://ww w.oldenbourg-verlag.de
Das W erk einschließlich aller A bbildungen ist urheberrechtlich geschützt. Jed e V erw ertung außerhalb der G renzen des U rheberrechtsgesetzes ist ohne Z ustim m ung des Verlages un zu
lässig und strafbar. Dies gilt insbesondere für V ervielfältigungen, Ü bersetzungen, M ik ro v er
film ungen und die E inspeicherung und Bearbeitung in elektronischen System en.
G edruckt auf säurefreiem , alterungsbeständigem Papier (chlorfrei gebleicht) G esam thcrstellung: R. O ldenbourg G raphische Betriebe D ruckerei Gm bH , M ünchen ISBN 3-486-56662-8
Inhalt
D a v i d C o h e n
Introduction ... VII Verzeichnis der Tagungsteilnehmer ... XI J o n E lster
Norms, Emotions and Social C ontrol ... I W illiam l a u M il le r
W eak Legs: Misbehavior before the E nem y ... 15 H e n k S. V ersnel
W riting Mortals and Reading Gods. Appeal to the Gods as a Dual
Strategy in Social Control ... 37 C h r i s t o p h e r A. F a r a o n e
Curses and Social C ontrol in the L a w Courts of Classical Athens ... 77 C y n t h i a P a t t e r s o n
The Polis and the Corpse: the R egulation of Burial in Democratic
Athens ... 93 G e r h a r d T h iir
Two ‘ C u rses’ from Mantineia (IPArk 8, IG V2, 262), Prayers for Justice,
and Oaths ... 109 K a r l - J o a c h i m H ö lk e s k a m p
Nomos, Thesmos und Verwandtes. Vergleichende Ü berlegungen zur Konzeptuahsierung geschriebenen Rechts im klassischen
Griechenland ... 115 A lb e r t o A4 a f f
Gesetzgebung und soziale O rdnung in Platons N omoi ... 147 J o c h e n M a r tin
to r m e n sozialer Kontrolle im republikanischen R om ... 155
VI Inhalt Lin Voxhall
Social Control, Roman Power and Greek Politics in the World of Plutarch ...
J o h n L. C o m a r o f f
Out of Control: an afterword ... jg g
D avid Cohen
Introduction
This volume of essays grew out of a conference on Democracy, Law, and Social Control at the Historisches Kolleg. The starting point for this conference was the question of how to understand the mechanisms of social control that operated in the particip atory dem ocracy of classical Athens. Since most previous approaches to this question have been quite narrow in focus, m y aim was to bring together an international group of leading scholars in law, classics, history, anthropology, and social theory w ho shared a common interest in different aspects of social control.
Since different disciplines have tended to approach social control from a variety of perspectives, the goal was to explore the w a y in which comparative studies can en
rich our understanding of processes of regulation, socialization, normalization, and deviance in traditional societies, and p articularly those of classical Greece and, to a lesser extent, Rome. The participation of classicists and ancient historians whose w o rk focuses upon these two cultures provided a central focus. At the same time, because the participants’ contributions addressed not only law, but also re
ligion, magic, gender, slavery, economy, war, and so on, w e were able to move beyond the narrow concentration of formal coercive institutions that has too often characterized studies in this area.
Even more importantly, the vital contribution of scholars whose w o rk involves other disciplines and historical areas provided us w ith comparative perspectives and theoretical models framing the general issues regarding social control that transcend any particular historical setting. The contribution of some of those scholars, such as Dieter Simon, Robert Bartlett, Peter Landau, and Christian Meier, was made through forms of participation other than publication of formal papers, but was nonetheless invaluable in stimulating the thinking of the authors represented in this volume. - Readers w ill note the absence of an index, w hich was felt to be unnecessary as the individual contributions are heavily thematic and analytical.
Athenian society of the classical era was a face-to-face society. In recent years some ancient historians (e.g. Osborne and Cartledge) have questioned this, ar
guing that historians like Moses Finley w ere naive in assuming that all Athenians knew one another and that social control operated in Athens in the direct w a y fa
miliar from studies of small village communities. Their doubts, however, rest upon the assumption that a face-to-face society can have no more than a few hun-
VI I I David Cohen
cired members, all of whom know one another intimately. As a review of the liter
ature of M editerranean anthropolo gy reveals, w hat matters instead are the q u a l i t a t i v e characteristics of social interaction. It is these, of course, which distinguish social control in Athens from anonymous modern mass urban societies.
Athens was a face-to-face society, but one which differed in important respects from the smaller communities which have been the object of so m any anthropo
logical studies of social control. N ot only did Athens posses a large and stratified population composed of citizens, slaves, metics, and foreigners, but it also exhibit
ed the complex centralized governmental and regulatory institutions of a state, albeit of a rather peculiar kind from the perspective of a modern state. Thus, al
though there were law courts, there was no state controlled coercive mechanism for the investigation or prosecution of crime, or for the enforcement of judgments in private lawsuits. Although there were state officials who were appointed to regulate m an y aspects of economic and public life, they differed radically from the modern bureaucracies whose discip linary and normalizing activities have been studied bv contem porary scholars of deviance and social control.
Above all, Athens was also a particip atory democracy. O n ly a m inority of the Athenian population w ere full citizens, but the participatory nature of Athenian law and politics makes it difficult to distinguish, for example, between official and unofficial, judicial and extra-judicial mechanism of social control. It was for this reason that I selected the title „Democracy, Law, and Social C o n tro l“ for the con
ference. The challenge is to understand the w a y in which social control operates in that particular type of direct, particip atory dem ocracy that we find in classical Athens. This was a dem ocracy centrally grounded upon an ideology of equality, but w hich also privileged the w ealth elite in crucially important w ay s and exclud
ed the m ajority of free inhabitants, to say nothing of slaves, from the exclusive male citizen club which alone governed. It is unfortunate that origin ally planned contributions on the most important disenfranchised groups, w om en and slaves, could not be included here, but it is w o rth noting the importance of extending the stu d y of social control to the w a y it operated for these groups who categorically did not participate.
As I suggested above, though the city of Athens was inhabited b y a fairly large scale urban society, it was not so large so as to exclude, in qualitative terms, those forms of social k now ledge and face-to-face interaction which distinguish it from a modern metropolis. As Aristotle fam ously argues in his P o li t i c s it was part of the defining characteristic of a Greek polis to be just such a face-to-face community.
O n his account if a polis is too large it cannot em body the moral com m u nity which makes a polis a distinctive form of political community. Above all, a polis must not be so large that it cannot deliberate collectively. And it must not be so large that citizens cannot have the know ledge of one another which is necessary for effective participatory politics. So while Athens m ay have differed in im por
tant respects from modern or traditional v ery small-scale face-to-face co m m u ni
ties, it also displayed m any similarities in the patterns by which social norms are articulated, interpreted, enforced, and reproduced.
Introduction I X The papers in this volume broaden the context of the stu d y of social control beyond the realm of formal judicial mechanisms of coercion. Including often ne
glected fields like religion and magic, as well as reflections upon political and legal theory, they also bring to bear the experience of other societies w ith very different institutional structures and traditions. These, together w ith perspectives of an
thropology, philosophy, and social th eory represented b y other contributors, in vite us to think beyond conventional accounts of social control and, I hope, be
yond the temporal and spatial boundaries of classical Athens.
Verzeichnis der Tagungsteilnehmer
Prof. Dr. Robert Bartlett, St A ndrews
Prof. Dr. David Cohen, Berkeley, Cal. (Stipendiat des Historischen Kollegs 1997/98)
Prof. Dr. John Comaroff, Chicago, 111.
Prof. Dr. Jon Elster, N e w York, N.Y.
Prof. Dr. Christopher A. Faraone, Chicago, 111.
Prof. Dr. Lin Foxhall, Leicester
Prof. Dr. Karl-Joachim Flölkeskamp, Köln Prof. Dr. Peter Landau, M ünchen
Prof. Dr. A lb erto Maffi, Mailand Prof. Dr. Jochen M artin , Freiburg i. Br.
Prof. Dr. Christian Meier, Hohenschäftlarn Prof. Dr. W illiam Ian Miller, A nn Arbor, Mich.
Prof. Dr. Dieter Nörr, München Prof. Dr. Dieter Simon, Berlin
Prof. Dr. C y n th ia Patterson, Atlanta, Ga.
Prof. Dr. Gerhard Thür, Graz Prof. Dr. H e n k Versnel, Leiden
Jon Elster
Norms, Emotions and Social Control '
I. Introduction
The term “social control” is m u ltiply controversial, (i) It suggests that “s o ciety” is capable of acting to “co ntrol” the behavior of its members. Yet societies cannot act: only individuals can. (ii) It is often also taken to suggest that the purpose of control is a benign one, that of reducing the extent of behavior that is in some sense harmful to society. Yet some mechanisms of social control are capable of causing harm as well as of containing it. (iii) The term suggests, finally, that the be
havior that is to be controlled is independent of the control mechanisms. Yet these mechanisms can generate the v ery behavior they regulate. T h ey can serve as p yro- maniacs as well as firefighters, puttin g out the fires they have set. N ote the differ
ence between (ii) and (iii). According to (ii), suppression of behavior through social control m ay be harmful. According to (iii), harmful behavior suppressed by mechanisms of social control m ay itself be caused by these mechanisms.
In the older literature on social control, these implied “suggestions” w ere part and parcel of the research paradigm. O n the one hand, these writings tended to vi
olate the principle of methodological individualism. O n the other hand, they often fell into an unthinking functionalism according to w hich the needs of “s o ciety”
create their own satisfaction. In the more recent literature, these suggestions are less pro m in ent1. Yet the very term “social control” has an insidious tendency to guide the mind along the lines of (i) - (iii) above. It w o u ld be better if one substi
tuted “mechanisms of social co ntrol” for “social control”, and even better if one simply talked about “social n o rm s”.
I am assuming, in fact, that social norms provide the mechanism of social con
trol. B y this assumption I exclude the l a w as a vehicle of social control. This is a purely stipulative definition, and hence cannot be defended b y argument. Yet I am prepared to argue that the differences between social norms and legal systems are so numerous and im portant that for most purposes it w o u ld be unhelpful to group
This paper draw s h eavily on m y “A lchem ies of the M in d ”, forthcom ing from C am bridge U niversity Press.
1 For an explicit rejection, see: R. B a r t l e t t , Trial b y Fire and W ater (O xford 1986) 34-42, w ho also provides some sam ples of the older literature. For other sam ples, see m y: N orm s of revenge, in: Ethics 100 (1990) 862-85.
2 Jon Elster
them together, either as explananda or as explanantia. Although legal and social norms have in common that both are enforced b y sanctions, they differ in that legal sanctions are formal, carried out b y specialized enforcers w ho have a clear self-interest in enforcing the sanctions (they would lose their jobs if they failed to do so). Social sanctions, b y contrast, are informal (there is no code stipulating the appropriate level of disapproval or ostracism for spitting in the street or rate-bust
ing); their enforcement is diffused am ong the population at large; and it’s not clear that the enforcers have an y interest in doin g w hat they do. Also, whereas one can assert (albeit not withou t some conceptual problems) that laws express the w ill of
“so c ie ty ” and that they tend to protect its members against harmful behavior, there is no basis for m aking similar claims about social norms. Finally, whereas so
cial norms are in tim ately linked to the emotions, laws and their enforcement need not owe anything to the latter.
The distinction between legal and social norms is not entirely hard-and-fast. In some cases, a moral norm against doing X gives rise both to a legal ban on X and to a social norm against X. Violence against children is an example. Also, a legal norm can generate a social norm. If there is a legal ban on driving faster than 60 miles per hour, some people w ill express disapproval of those w ho exceed that speed. Conversely, some legal norms m a y sim p ly formalize a preexisting social norm. Social norms against incest antedate laws against incest. Yet all these examples involve benign norms, w hich in some obvious sense benefit the members of society. Social norms that suppress socially useful behavior or gener
ate socially harmful behavior are less lik ely to be encoded in law. Also, as noted below, laws m ay be necessary because the appropriate norms are lacking.
I now proceed as follows. In Section II I consider the reciprocal relation be
tween emotions and social norms. In Section III I turn to the behaviors that are prescribed or proscribed b y social norms. Section IV offers a brief conclusion.
A lthough the main thrust of the paper is general, the more extensive illustrations are taken from ancient Greece.
II. Em otions and social norm s
The relation between emotions and social norms is twofold. M ost obviously, the emotions of shame and contempt ensure the efficacy of social norms. A t the same time and som ew hat more paradoxically, emotions (along with other motivations) can themselves be the target of social norms.
E m o t i o n s as t h e s u p p o r t o f s o c i a l n o r m s . Social norms as I understand them are non-outcome-oriented injunctions to act. In their simplest form, they take the form of unconditional imperatives: “alw ays w ear black at funerals”. T h ey m ay be contrasted w ith outcome-oriented imperatives: "alw a ys wear black in strong sun
shine” (as do people in Mediterranean countries to maintain circulation of air be
tween the clothes and the body). In a more complex form, they can be conditional
Norms, Emotions and Social Contro l 3 imperatives that make the action contingent on the past behavior of oneself or others rather than on future outcomes to be achieved. N orms of reciprocity, for instance, have this form: help those w ho help yo u , and harm those who harm you.
For these norms to be s o c i a l rather than m erely private rules, they have to be shared w ith other members of the relevant culture or subculture.
Social norms affect action through the disapproval meted out to norm-viol- ators. There are tw o distinct mechanisms involved. On the one hand, disapproval often goes together w ith material sanctions. One can refuse to have any dealings with a norm-violator, or punish him directly in a variety of w ays. For some writers, this is the main w a y social norms operate2. O n the other hand, disap
proval often induces painful feelings in the norm-violator. Two correlative em o
tions are involved: contempt in the observer of the norm-viola tion and shame in the norm-violator. As Aristotle noted, the deterrent effect of shame must be dis
tinguished from the deterrent effect of the concomitant material sanctions.
“Shame is the imagination of disgrace, in w hich we shrink from the disgrace itself and n o t f r o m its c o n s e q u e n c e s ” (Rhetoric 1384a; italics added).
In m y opinion, the emotional sanctions are more important than the material ones. In fact, the latter matter m ain ly in so far as they are vehicles for the former.
W hen I refuse to deal w ith a person who has violated a social norm, he m ay suffer a financial loss. M ore importantly, however, he will see the sanction as a vehicle for the emotions of contempt or disgust, and suffer shame as a result. The material aspect of the sanction that matters is h o w m u c h it co s t s t h e s a n c t i o n e r to p e n a l i z e the target, not h o w much it costs the target to be penalized. The more it costs me to refuse to deal w ith him, the stronger he will feel the contempt behind m y re
fusal and the more acute w ill be his shame. A lthough high costs to the sanctioner often go together with high costs for the target, as when the sanctioner renounces the opportunity for a m u tu ally profitable business transaction, this need not be the case; and even w h en it is the case, m y claim is that the costs to the sanctioner are what makes the sanction really painful to the target. It tells him that others see him as so bad that they are willing to forego valuable opportunities rather than have to deal w ith him.
Besides shame, the emotion of g u i l t is also capable of modulatin g behavior. A l
though both emotions are probably found in all societies, their relative im p o r
tance differs. Ancient Greece, for instance, was closer to the shame end than to the guilt end of the continuu m 3. For the Greeks, “goodness divorced from a repu-
1 D. A b re u , On the theory of infin itely repeated games w ith discounting, in: E conom etrica 56 (1988) 3 83-96; G. A k e r l o f, The econom ics of caste and of the rat race and other w oeful tales, in: Q u arterly Jo u rn al of Economics 90 (1976) 599-617; R. A x e l r o d , An evolution ary approach to norm s, in: A m erican P olitical Science R eview 80 (1986) 1095-1111;/. C o l e m a n , foundations of Social T h eo ry (C am bridge, M ass. 1990).
3 D. C a i r n s , A idos: The P sych o lo gy and Ethics of H onour and Shame in A ncient G reek L it
erature (O xford 1993) 27-47; B. W illia m s, Sham e and N ecessity (B erkeley 1993); K. D o v e r , Greek Popular M o rality (Indianapolis 1994) 220-23.
4 Jon Elster
tation for goodness was of limited interest”4. Conversely, what th ey feared was being s e e n to act badly rather than the bad action itself. This being said, the Greeks carved out a conception of guilt w ithin their conception of shame. Bernard W il
liams observes the Greeks were concerned w ith m an y of the things that w e associ
ate w ith guilt rather than shame, such as indignation, reparation and forgiveness3.
There is also something like a concept of guilt in A ristotle’s discussion of shame.
He writes that w e are not “ashamed of the same things before intimates as before strangers, but before the former of w hat seem genuine faults, before the latter of w hat seem conventional ones” (Rhetoric 1384 b). Also, he notes, w e feel friendly
“towards those w ith w h o m w e are on such terms that, while we respect their opinions, w e need not blush before them for doing w hat is conventionally wrong;
as w ell as towards those before w h o m we should be ashamed of doing anythin g really w r o n g ” (1381 b). Something like the shame-guilt distinction appears here, then, as a distinction between shame before strangers and shame before friends, or, equivalently, between shame for conventionally w ro n g actions and shame for g enu inely w ro n g actions6.
Even in societies that have a more prominent place for guilt, shame provides the stronger motivation. A .O . L o vejoy quotes Voltaire as saying that “To be an object of contempt to those w ith w h o m one lives is a thing that none has ever been, or ever w ill be, able to end u re”, A dam Sm ith as asserting that “Com pared with the contempt of mankind, all other evils are easily sup p orted ”, and John Adams to the effect that “The desire of esteem is as real a w ant of nature as hunger; and the ne
glect and contempt of the w orld as severe a pain as gout and stone”7. M odern p s y chologists also assert that the burning feeling of shame is more intensely painful than the pang of guilt8. To m y knowledge, people rarely commit suicide out of guilt. B y contrast, six Frenchmen killed themselves in 1997 after they were caught in a crackdow n on pedophilia.
Feelings of shame must have been u n u su ally strong in the competitive world of the Greeks. Aeschylus, for instance, “w rote his plays for performance at a dra
matic competition w ith the hope presum ably of securing first p r iz e ”; hence when '* D o v e r , G reek P opular M o rality 226. We m ay contrast this attitude w ith that of M on ta ig n e-.
“The more glittering the deed the more I subtract from its m oral w orth, because of the su s
picion aroused in me that it w as exposed m ore for glitter than for goodness: goods d isplayed are alread y h alfw ay to being so ld .” (The C om plete Essays [H arm ondsw orth 1991] 1157f.) S im ilarly, P a s c a l w rote that “the finest things about [fine deeds] w as the attem pt to keep them secret” and that "the detail b y w hich they cam e to light spoils ev eryth in g ” (Pensees, S e llie r [ed.], 520).
5 W illiam s, Sham e and N ecessity 91.
6 In his discussion of “conscience” D o v e r , G reek P opular M o rality 220-223 b y and large agrees that the G reeks did not have our (no n -religio us) concept of guilt, but also cites “pas
sages w hich seem to carry a suggestion (perhaps in some cases illu so ry) that self-respect and the prospect for self-contem pt are genuine m o tives” (221).
7 A.O. L o v e j o y , R eflections on H um an N ature (B altim ore 1961) 181, 191, 199.
8 M. L ew is , Shame (N ew York 1992) 77; J . P T a n g n e y , A ssessing individual differences in proneness to shame and guilt: developm ent of the self-conscious affect and attrib ution inven
tory, in: Jo urn al of P erso n ality and Social P sych o lo gy 59 (1990) 102-11, at 103.
Norms, Emotions and Social Control 5 he left Athens for Sicily “it will occasion no surprise that one reason advanced in antiquity for his departure from Athens was professional chagrin, defeat at the hands of the yo u n g Sophocles or at the hands of Sim onides”9. A m ong the Greeks, losing was alw ays shameful; in fact “defeat is a i s c h r o n [shameful] even when the gods cause it” 10. The value attached to glo ry in the Greek w orld was so strong that shame could attach not only to losers but also to non-contestants. Thus in X eno
p h o n ’s M em orabilia (III.7.1) Socrates and Charmides agree that “a man who was capable of gaining a victory in the great games and consequently of winning hon
our for himself and adding to his co u n try ’s fame, and y e t refused to com pete”
could o n ly be a coward. Moreover, as I argue below, feelings of shame were inten
sified b y the fact that n o body bothered to hide their contempt.
E m o t i o n s as t h e t a r g e t o f s o c i a l n o r m s . Typically, social norms prescribe or pro
scribe specific forms of b e h a v i o r . Social norms targeting emotions w o u ld not seem to make much sense. In the standard case, observed violation of a norm triggers a sanction that provides an incentive to respect the norm. Emotions, however, are largely unobservable; also they cannot be modified b y incentives. It is nevertheless indisputable that people can feel ashamed of their emotions or lack of emotion.
O ne is supposed to be h appy at one’s wedding d ay or sad at a funeral, and failure to experience these emotions is likely to induce shame. Because of strong social norms against envy people often feel ashamed when th ey experience this emotion.
In societies that value courage very highly, people m ay be ashamed of being afraid.
If emotions w ere nothing but in voluntary and unobservable mental states it w ould indeed be paradoxical to have norms targeting them. Yet emotions also have physiological and behavioral e x p r es si o n s , w hich are often observable and at least p artly under the control of the will. Some of the em otion-related norms do in fact only target the expression of emotions. These “display ru les”, as Paul Ekman calls them, are m ain ly rules of etiquette11. Just as there is a norm enjoining me to wear black at a funeral, there is one that tells me to put m y face in serious folds even if I was not p articularly close to the deceased. The behavior is intended to show respect for those who grieve, not to make them (or others) believe that I am grieving. In other cases, a person m ay simulate emotional expressions (or hide them) in order to make others think that he is (or isn’t) in a specific emotional state. One reason for doin g so is sq uarely instrumental: I m ay fake anger, to get m y w ay; or hide m y anger, to avoid attack. Another reason is provided by the need to abide by a social norm. I m ay try to hide m y fear or m y envy because of the opprobrium often attached to these emotions, or simulate grief w hen I w a s close to the deceased.
9 P. Walcot, E nvy and the G reeks (W arm inster 1978) 5 0 f.
Iu A. W. H. Adkins, M oral Values and P olitical B ehaviour in A ncient Greece (N ew York 1972) 60.
11 P. E k m a n , Biological and cultu ral contributions to b ody and facial m ovem ent in the ex
pression of the em otions, in: A. R o r t y (ed.), E xplaining the Em otions (Berkeley, Los Angeles 1980) 73-102.
I w o u ld like to place the issue in a larger perspective, b y considering the general phenomenon of norms that are directed towards the motivation behind behavior rather than towards the behavior itself. For this purpose I shall use the tripartite classification of emotions suggested by La B ruyere: “N othing is easier for passion than to overcome reason; its greatest trium ph is to conquer interest” (Characters IV.77). I shall understand the ideas of passion (or emotion) and interest more or less in their e v eryd ay sense. As for reason, I shall understand it as an y kind of impartial (disinterested and dispassionate) motivation, aimed at prom otin g the public good, individual rights, and the like.
In an y given society, these motivations (and their subspecies) will appear in a normative hierarchy. Some will be valued highly, others tolerated, still others des
pised. These evaluations set up a pressure on individuals to misrepresent their motivations to others so as to appear in a more favorable lig h t12. For illustration, consider the M elian dialogue in which the Athenians make the following state
ment to the Melians:
F or ourselves, we shall not trouble you w ith specious pretenses - either of how we have a right to our em pire because we overthrew the M ede, or arc now attacking you because of w rong that you have done us - and make a long speech w hich w ould not be believed; and in return we hope that yo u , instead of thin kin g to influence us b y sayin g that yo u did not join the Spartans, although their colonists, or that you have done us no w rong, w ill aim at w hat is feasible, h olding in view the real sentim ents of us both; since you know as w ell as W 'e do that right, as the w orld goes, is o nly in question betw een equals in power, w hile the strong do w hat they can and the w eak suffer w'hat they must. (5.89)
Flere, the Athenians very explicitly claim that they w ill n o t tr y to misrepresent their real motives. C om m enting on this and other passages, A. H. M. Jones writes that “If these speeches are intended to reproduce the actual tenor of Athenian public utterances, it must be admitted that the Athenians of the fifth century were [ .. .] a v ery remarkable, if not unique, people in adm ittin g openly that their p o lic y was guided purely by selfish considerations and that th ey had no regard for p oliti
cal m o ra lity ”. In fact, he finds the speeches so implausible that he concludes that
“Thucydides, in order to point his moral, put into the mouths of Athenian spo
kesmen w h at he considered to be their real sentiments, stripped of rhetorical clap
trap ” 13. Even the notorio usly frank Greeks, in other words, w o u ld not be that frank. In other cultures, presumably, the pressure towards misrepresentation w ould be even stronger.
In ancient Greece, the unav ow ab ility of envy, h y b r i s and interest induced com moners as w ell as kings to present actions thus motivated in a different light. Plu
tarch writes that “men d en y that they envy [...]; and if you show that they do, they allege an y num ber of excuses and say they are an gry with the fellow or hate 12 T h ey also set up a pressure on individuals to m isrepresent their em otions to t h e m s e l v e s , as w hen an envious person redescribes the situation so as to ju stify the more acceptable emotion of righteous indignation. For reasons of space, I ignore this phenom enon here. For reasons that I do not understand w ell, it does not seem to be prom inent in the w ritin gs of the Greeks.
13 A. H. M. J o n e s , A thenian D em ocracy (B altim ore 1957) 66 f.
him, cloaking and concealing their envy with whatever other name occurs to them for their passion” 14. In classical Athens, this tendency was revealed by the practice of denouncing others, w ho claimed to act for the sake of revenge, of being really motivated b y envy. In David C o h e n ’s summary, the orator Lysias “argues that his opponent will falsely claim that he brings the prosecution out of enm ity so as to get revenge, but in fact it is o n ly out of envy because the speaker is a better citizen, [...] The desire lo r revenge apparently w ould be seen b y the judges as a legitimate reason for prosecuting, so the speaker must d en y that this is the case. Meanspirited envy, on the other hand, reflects b adly upon the accuser’s character and indicates that the suit is unreliable.” 1-'’
In Athens, h y b r i s was a punishable offense and, moreover, a strongly disap
proved form of behavior. There was a legal category, g r a p h e h y b r e o s , w hich en
abled victims (or others) to prosecute hybristic behavior. Again st accusations of h y b r i s therefore, it was expedient to represent one’s behavior as motivated b y a more acceptable urge. In the Politics (1311 b), Aristotle tells a story about a tyrant, Archelaus, w ho was killed (among other reasons) because one of his boyfriends decided that their association had been based on “h y b r i s not on erotic desire” . Aristotle then offers the advice to tyrants w ho w an t to stay in power, that in their acquaintances w ith youth, they should appear to be acting from desire rather than from h y b r i s 16. In other contexts, those accused of h y b r i s represent their behavior as motivated b y revenge. A lthough there m ay have been truth in their allegations of having been wronged, th ey m ight still be gu ilty of h y b r i s if the revenge was dis
proportionate to the offense17.
In classical Athens, there were “sycop h an ts” or professional accusers w ho initi
ated public lawsuits (g r a p h e ) for private gain, either because they could hope for a share of the fine or because they hoped that even innocent plaintiffs w o u ld settle in private rather than taking the risk of litigation. As sycophants were regarded with deep suspicion, it was important for them to misrepresent their motivation.
As explained by Mogens H erman Hansen, it was more effective to disguise their interest as passion than to try to pass themselves off as motivated b y impartial mo
tives: “W hen a citizen appeared in court as a public accuser his first anxiety was [...] to dispel any suspicion that he was a sycophant. H e could stress his public
spiritedness, but that tends to make ordin ary folk even more suspicious, and usually there was a much more cogent argument to deplo y: he could declare that the accused was his personal enem y and that he was using his citizen right to pros
ecute for revenge and not for gain.” 18 14 P lu ta r ch , On envy and hate.
b D. C o h e n , Law, Violence and C om m un ity in C lassical A thens (C am brid ge 1995) 82 f.
16 Politics 1315 a; see also C o h e n , Law, Violence and C o m m un ity in C lassical Athens 145 and N. R. E. Fisher, H yb ris (W arm inster 1992) 3 0 f.
17 Fisher, H yb ris 509 (sum m arizing his analyses in earlier chapters, n otab ly C h. XI).
18 M. H. H a n s e n , The A thenian D em ocracy in the Age of D em osthenes (O xford 1991) 195;
see also:/. O b e r , M ass and Elite in D em ocratic A thens (Princeton 1989) 212 for a sim ilar comment.
8 Jo n Elster
Innumerable speeches of the Greek orators make it clear that the desire to en
hance the glo ry of the polis was at the apex of the hierarchy of motivations. As the passage just cited implies, claims to be public-spirited w ere not alw ays taken at face value. Dover also notes that the category of “p o ly p ra g m o sy n e ” or meddle
someness was used about “the man w ho claims alw ays to be read y to prosecute in the c ity ’s in terest” 19. Yet this fact does not exclude - on the contrary, it presup
poses - the p rim acy of public-spiritedness. The other passages I have cited suggest that in the hierarchy of motivations, reason was ranged above revenge (a passion), w hich was above interest, which was above envy and hybris (also passions). Of the latter tw o, envy seems to have been viewed as the most despicable. In spite of the legal ban on h y b r i s , an Alcibiades fascinated his fellow-citizens in a w a y that an envious person could never do.
A n important feature of the Greeks can be sum m arized b y saying that they w ere (relatively) u n a s h a m e d o f s h a m i n g . R obert Levy writes that in Tahiti, there is both control b y shame (h a ’a m a ) and control of shame: “A lthough gossip is an im portant part of ,shame co ntrol’, the w ords designating gossip have a pejorative tone, and gossiping is said to be a bad thing to do. Ideally, the behavior which w o u ld produce shame on becoming visible has to spontaneously force its w a y into visibility; people are not supposed to search out shameful acts. Such a searching out is itself a h a ’a m a th ing.”20 Ancient Greece, by contrast, was a w o rld with v ery little emotional tact, a w orld in which a man was not afraid to express disapproval of others m erely because they were born u g ly or poor - just as a child in our own society m ay express spontaneous disgust at the sight of a disfigured person. Ober cites a law “that forbade anyone to reproach an y Athenian, male or female with w o rk in g in the ag o ra”21, presupposing both a tendency to disapprove of such w o r k and a tendency to disapprove of the disapproval. In modern Western so
cieties, the latter is sufficient to neutralize the former, either because the one is v ery strong or because the other is v ery weak. A m o ng the Greeks, the relative strength of the tw o tendencies was such that a law was needed.
III. Social norms and behavior
Writers on social norms tend to assume that they are by and large utilitarian, either in the sense of benefiting all members of society or in the sense of benefiting some at the expense of others22. M a n y also tend to assume that the benefits ex p la in w h y the norms exist23. C o n tra ry to these writers, I believe there are norms that do not
19 D o v e r , P opular G reek M o ra lity 188.
20 R. L ev y , The Tahitians (C hicago 1973) 340.
21 O b e r , M ass and E lite in D em ocratic A thens 276.
22 For a more refined classification, see: /. C o l e m a n , The Foundations of Social T heory (C am bridge, M ass. 1990) 246-49.
23 A gain I refer to C o l e m a n (ibid., C h. 11).
Norm s, Emotions and Social Control 9 benefit anyone, and that even w hen a norm does provide benefits one cannot as
sume w ithou t further argum ent that these have explanatory power. Here I shall focus on the behavioral effects of norms, leaving the explanatory issues aside24.
N o r m s t h a t b e n e f i t e v e r y o n e . It is undeniable that some social norms w o rk out to the benefit of everybody. The norm against spitting in public has (or had) beneficial hygienic effects. The norm against incest reduces the number of persons born w ith genetic defects. The norm of voting in general elections sustains dem oc
racy, to e v e ry b o d y ’s benefit. N orms against envy give more room for innovation and entrepreneurship. N orm s against vengeance reduce overall levels of violence.
At a more general level, the norm of reciprocity w o rks out to the benefit of all, at least when combined w ith the norm that others should be trusted if and only if they have not show n themselves to be u n w o rth y of trust ( “Fool me once, shame on you; fool me twice, shame on m e ”). •
In the Greek context, it is at least arguable (but h ard ly provable) that the strict n o r m s o f a c c o u n t a b i l i t y m ay have w orked out to e v e ry b o d y ’s benefit in the long run, even if on a given occasion they m ay have led to m anifestly unjust out
comes25. The Athenian mode of accountability did in fact resemble the legal principle of strict lia bility26. Theirs was not only a “s h am e-cultu re”, but also a
“results-culture”27, in w hich people were held accountable for the outcome of their actions regardless of m itigating or extenuating circumstances28. If this p rin
ciple was adopted in Western societies today, it w o u ld pro b ab ly have undesirable results on the whole. In a society that was almost constantly at war, such as the Athenian democracy, it m ay have been useful overall. As Dover notes, “a nation at war turns itself into an organization with a specific and definable purpose, and it deals more severely w ith negligence and inefficiency than a nation at peace; the more perilous its situation, the less im portance is attached to distinctions between incapacity, thoughtlessness and treachery.”29 I leave it to the reader to ponder whether the perilous situation i n d u c e d the adoption of the results-culture, or whether this attitude was sim p ly a useful b y-product of the shame-culture.
N o r m s t h a t b e n e f i t s o m e a t t h e e x p e n s e o f o t h e r s . Some norms have a utilitarian aspect b y favoring some members of society at the expense of others. In contem
porary Western societies, the norm against smokin g in public benefits (or is be
lieved to benefit) non-sm okers, at the expense of smokers. N orms of equality serve the interest of those w ho are badly off at the expense of those w ho are well off. In hierarchical societies, norms of deference serve the upper tiers of the social
24 I discuss those issues in: The C em ent of Society (C am brid ge 1989) 147-49.
25 The follow ing draw s on m y: A cco u n tab ility in A thenian p olitics, in: B. M a n in , A. P r z e - 'w o rsk i, S. S to k e s (eds.), D em ocracy, A cco u n tab ility and R epresentation (forthcom ing from C am bridge U n iv ersity Press).
26 Williams, Sham e and N ecessity, C h. 3.
27 Adkins, M oral Values and P o litical Behaviour in A ncient G reece 61.
28 For other exam ples of stric t-lia b ility societies see: R. E d g e r t o n , R ules, Exceptions and the Social O rder (Berkeley, Los Angeles 1985) 161 f.
29 D o v e r , Popular G reek M o rality 159.
10 Jo n Elster
system at the expense of the lo w er ones. I assume that the norms are held b y those w hom they harm as w ell as by those w hom they benefit. Hence the statement
“Children should be seen and not h eard ” does not, in m y terminology, express a social norm unless it is one to w hich children also subscribe30. When adults en
force this principle m erely through their power to punish children w e are dealing w ith a very different phenomenon from w hat we observe when members of a sub
ordinate class police each other to ensure the proper deference to their superiors.
In the latter case, but not in the former, emotions also come into play.
M a n y Greek writers and their modern commentators have asserted a strong be
havioral impact of norms of eq uality in such matters as ostracism, liturgies and j u r y sentencing31. Yet in most cases, alternative explanations are possible. O stra
cism could be used som ew hat like our general elections, to decide between tw o al
ternative policies b y expelling the proponent of one of them. To the extent that it was directed against individuals rather than against policies, the motivation is not necessarily one of envy. Alcibiades, for instance, was w id e ly seen to be hybnstic, and the fear that he might use pow er to set up a ty r a n n y was not at all im plau
sible32. As Paul Veyne has argued, convincingly to m y mind, the Greek system of liturgies was based on more complex motivations than fear of egalitarian envy33.
A nd if the Greek juries imposed heavy fines on rich defendants, the motive m ay have been interest rather than envy. In a hand-to-m outh econom y such as Athens there was alw ays “a temptation to jurors to vote in the interest of the treasury w hen m oney was short, and an informer dangled before their eyes a fat estate w hose owner, he alleged, had been guilty of some serious offense”34. More gen
erally, the Athenians w ere much too susceptible to norms of wealth and status for any simple egalitarian view of their society to make sense35.
30 H ere m y term ino logy differs from that of: C o l e m a n , F oundations of Social T heory 247, from w hom I take this example.
31 W alco t, E nvy and the G reeks, is a useful sum m ary.
32 Fisher, H yb ris 87.
33 P. V eyn e, Le Pain et le cirque (Paris 1976) C h. II.
34 J o n e s , A thenian D em ocracy 58, citing three speeches by Lysias (30, 27, 19). A proposal by A ristotle w ould , if im plem ented, have provided an ingenious w ay of testing the envy h y pothesis versus the interest hypothesis. In the P olitics, he notes that “The dem agogues o f o u r own day often get p ro p erty confiscated in law -co urts to please the p eo p le” (1320 a), because the people has a direct financial interest in the size of the state coffers. H e then goes on to rec
om mend “a law that the p ro perty of the condem ned should not be public and go into the treasury but be sacred. Thus offenders w ill be as much afraid, for they w ill be punished all the sam e, and the people, having nothing to gain, w ill not be so ready to condem n the accused”.
In terms of La B ru y ere ’s trich o to m y of m otives, A risto tle claim s that if you rem ove any i n t e r e s t the people m ight have in the outcom e, they w ill decide in accordance w ith r e a s o n or justice, thus assum ing that they w ou ld not be moved by p a s s i o n , e.g. by envy.
35 O b e r , M ass and Elite in D em ocratic A thens 224 f., 2 8 7 f. H is argum ent is based on the fact that the classical orators offered elitist argum ents w hen addressing the dem ocratic jury.
D o v e r , G reek P opular M o rality 3 4 f. discusses w hether the jurors w ere prosperous or sim ply liked to be treated as if they w ere, and opts, tentatively, for the form er alternative. O b e r ar
gues (op.cit., p .141) that the juro rs w ere representative of the population as a w hole; see also:
N orm s, Emotions and Social Contro l 11 N o r m s t h a t b e n e f i t n o o n e . I have discussed norms that benefit all and norms that benefit some at the expense of others. But some norms do not benefit a n y one36. Consid er the norm in our society against w alk in g up to the person at the head of the bus queue and asking to b u y his or her place in the queue. This prac
tice, if allowed, w o u ld not harm anyone. The person asked to give up the place is free to refuse. If the offer is accepted both parties to the transaction will be better off and no third parties w ill be hurt. B y blocking such potential Pareto-improve- ments, the norm makes e veryb o d y worse off. Or consider the pointless suffering induced by norms of etiquette, which penalize people for wearing the w ro n g kind of clothes or having the wro n g kind of haircut. The argum ent that these norms are useful in that adherence to them “w ill declare o ne’s group id entity to other members and to nonm em bers”37 m ay be adequate in some cases, but hard ly in all.
W hen a small girl comes home c rying because her friends ridicule her purchase of the w rong sort of pram for her doll, no useful function is served. These m ay seem to be inconsequential matters, and in one sense they clearly are. Yet as Tocqueville noted, although “nothing, at first sight, seems less important than the external formalities of h um an behavior [...], there is nothing to w hich men attach greater im portance”38. Proust and Edith W harton w o u ld have concurred.
Some norms that are un am b iguo usly consequential also fail to provide any benefits. In m y view, norms of revenge fall in this category39. The Mediterranean and M iddle Eastern societies that subscribe to these norms have levels of violence and m ortality rates among yo u n g men far above w hat is found elsewhere. The idea that the practice of revenge is a useful form of population control is too arbitrary to be taken seriously. The idea that norms of revenge provide a functional equiv
alent of organized law enforcement in societies w ith a w e ak state is also fallacious, albeit more subtly. N o rm s of revenge and the larger code of honor in w hich they are u sually embedded set, as I said in the opening paragraph, as m an y fires as they put out. In m any cases, the question “w hether feuds created more disruption than they controlled”40 m ay be answered in the affirmative.
In dueling and feuding societies, m any people do in fact engage in deliberate provocation, to insult or offend another. Moreover, one cannot achieve honor by insulting just anybody. In Iceland, "the possession of honor attracted challenges, because that was where honor was to be h a d ”41. For a medieval knight, the “prime concern must be pursuit of distinction, and a challenge should never be rejected.
M.H. H a n s e n , The A thenian D em ocracy in the Age of D em osthenes (O xford 1991) 186 (the poor and the eld erly w ere the m ajo rity in the courts).
This statem ent is slig h tly inaccurate. U sually, some individuals benefit ex p o s t from the operation of the norm s discussed below. This is com patible w ith the idea that nobody bene
fits ex a n t e , and a fortio ri w ith the idea that the average benefit is negative.
3/ C o l e m a n , Foundations of Social T h eory 258; also: P. B o u r d i e u , La distinction (Paris 1979).
38 D em ocracy in A m erica 605.
-’9 Ih e follow ing draw s on m y “N orm s of revenge”.
4u C. B o e h m , Blood Revenge: The A nthrop o lo gy of Feuding in M ontenegro and O ther tribal Societies (L aw rence 1984) 183.
41 W. I. M iller, B loodtaking and Peacem aking (C hicago 1990) 33.
12 Jon Elster
Rather he should go out of his w a y to confront others.”42 M ontaigne refers to
“w hat is said b y the Italians when they wish to reprove that rash bravery found in yo unger men by calling them b i s o g n o s i d ’h o n o r e , ,needy of ho n ou r’ : they say that since th ey are still h u n gry for that reputation, w h ich is hard to come by, they are right to go and look for it at an y price - something w hich ought not to be done by those w ho have already acquired a store of it.”43 The behavior of Meidias when he slapped Demosthenes in public during a festival conforms to this pattern.
Athenian society was based, ro u ghly speaking, on the pursuit of glory through competition and of honor through confrontation. M ore than anything, the A th e
nians wanted to excel and to stand out. Other societies have been based squarely on the norm against sticking one’s neck out. In Aksel Sandemose’s “L aw of J a n te ” (a m ythical small tow n in Denmark), the fourth of the ten commandments is
“Thou shalt not fancy thyself better than ^ e ”44. Keith Thomas writes that in m an y prim itive societies, beliefs in witch-craft “are a conservative force, acting as a check on undue individual effort. Similarly, in tw elfth-century England the chronicler W illiam M alm esb u ry could complain that the common people dis
paraged excellence in an y sphere b y attributing it to demonic a id .”45 U nlike egali
tarian norms that have a redistributive effect, the norm against sticking one’s neck out does not benefit anyone.
IV. C onclusion
I have argued elsewhere46 that the idea of “social o rd e r” can be understood in two distinct ways. O n the one hand, social order is the solution to the H obbesian d i
lemma, the achievement of the cooperative solution to a collective action problem.
O n the other hand, social order is the avoidance of chaos, the realization of a stable and predictable state of affairs. This state m a y be ab y sm a lly bad, but at least it offers no surprises.
The idea of “social control” m ay be taken in a sim ilarly dual sense. O n the one hand, the control m ay target individuals who engage in non-cooperative behavior.
O n the other hand, they m ay target individuals w ho deviate from expectations. In either case, the control m a y be achieved through the vehicle of social norms. Some norms promote cooperation b y punishing non-cooperators. Other norms p ro
42 V. K i e r n a n , T he D uel in E uropean H isto ry (O xford 1986) 33.
43 M o n t a i g n e , The C om plete Essays 839; R. F, B r y s o n , The P oint of H onor in Sixteenth- C en tu ry Italy (N ew York: P ublications of the Institute of French Studies, C olum bia U n iver
sity 1935) 28, cites a sixteen th-cen tury Italian w riter to the effect that “giving [insults] p er
tains to the nature of man; because everyone seeks d istinction, one m ark of which is to offend fearlessly”.
44 A. S a n d e m o s e , A F ugitive C rosses his Track (N ew York 1936) 77.
45 K. T h o m a s , R eligio n and the D ecline of M agic (F larm ondsw orth 1973) 644.
46 In the Introduction to: "The C em ent of S o ciety”.
Norms, Emotions and Social Control 13 mote stability b y punishin g deviants. And in either case, the norms operate through the social emotions of shame and contempt.
Classical Athens does not fit either case very well. T he most prominent norms involved competition and confrontation, w hich are constant-sum forms of inter
action. It was neither true that everybody gained b y the operation of the norms nor that nobody gained: rather, some gained at the expense of others. (At least this view can be defended for the competitive pursuit of glory. As noted above, the conf rontational pursuit of honor is more ambiguous.) To the extent that the A th e nians managed to solve their collective action problems, cooperation was achieved through the law rather than b y informal social norms. A t the same time, as David Cohen has shown, the legal system was itself a vehicle of confrontation and com petition47. For that reason, the distinction between social norms and legal systems that I made initially is more tenuous in Athens than in modern societies. Informal social norms pervaded even the sphere of legal regulation of behavior.
47 C o h e n , Law, Violence and C o m m un ity in C lassical A thens.
William Ian Miller
Weak Legs: Misbehavior before the Enemy*
Statutes make for appallingly tedious reading unless prim itively short and to the point as, for example, this provision in the early Kentish laws of /Ethelberht (c. 600): “H e w ho smashes a chin bone [of another] shall p a y 20 shillings” or this one from King Alfred (c. 890): “If anyone utters a public slander, and it is proved against him, he shall make no lighter amends than the carving out of his tongue.” 1 Yet on very rare occasion a modern statute can rivet our attention and when it does it seems to do so by mim icking some of the look and feel of legislation enacted in less law yer-ridden times. Consider the statute presently codified in the United States C ode as part of the U niform C ode of M ilita ry Justice:
Misbehavior before the enemy
A n y member of the armed forces who before or in the presence of the enemy:
( ! ) runs aw ay;
(2) shamefully abandons, surrenders, or delivers up an y command, unit, place, or military property w h ich it is his d u ty to defend;
(3) through disobedience, neglect, or intentional misconduct endangers the safety of an y such command, unit, place, or m ilitary property;
(4) casts a w a y his arms or ammunition;
(5) is guilty of c o w a rd ly conduct;
(6) quits his place of d u ty to plunder or pillage;
(7) causes false alarms in an y command, unit, or place under control of the armed forces;
(8) w illfully fails to do his utmost to encounter, engage, capture, or destroy any enemy troops, combatants, vessels, aircraft, or an y other thing, w hich it is his duty so to encounter, engage, capture, or destroy;
(9) does not afford all practicable relief and assistance to an y troops, combatants, vessels or aircraft of the armed forces ... when engaged in battle;
shall be punished by death or such other punishment as a court-martial m ay direct2.
Spccial thanks to L a rry Kramer. - P ortions of this essay are included in 'William J a n Miller, I he M ystery of C ourage (C am bridge, M A 2000); and an earlier version appeared in: R ep re
sentations 70 (2000) 27^-48,
1 / E thelberht cap. 50; A l f r e d cap. 32. The provisions are most conveniently accessible in: E L.
A t t e n b o r o u g h (ed.), The L aw s of the E arliest English Kings (C am brid ge 1922) 11, 77. I have altered the diction of A tten b orou gh ’s translation.
2 10 U SC S @ 899 (1997) A rt. 99.
16 William Ian Miller
M ak in g cowardice a capital offense strikes us as a kind of barbaric survival from a rougher age, a time, that is, when few doubted that courage ranked higher than pitv or prudence in the scale of virtues. And if many of us to day believe that capi
tal punishment cannot be justified even for the sadistic torturer, w hat a shock to discover that, as an official matter at least, Congress reserves it for the person who cannot kill at all. N ot to w o rry : although the state has the power and right to execute those w ho misbehave before the enemy we are too unsure of ourselves, or m aybe even too charitable, to enforce the statute maximally. We have done so but once since 1865 when Private Eddie Slovik was executed by firing squad ‘pour en- courager les autres’ in the bleak Eliirtigen Forest of 19453. Still, even if only by inertia, we have preserved the option.
Quite independent of the grimness of its sanctions, the statute prompts our attention because of its strangely absurdist quality. M ost of its provisions seem m erely to restate each other. What, for instance, is running a w a y (1) that isn ’t also co w ard ly conduct (5). A nd aren’t paragraphs 2 and 8, the one covering the shame
fulness of cowardice on defense, the other governing slacking off on offense, really special cases of c o w a rd ly conduct punished in 5? Paragraph 7 goes so far as to make jitteriness a capital offense to the extent on e’s nerves lead one to overin
terpret causes for alarm, w hile paragraph 3, in contrast, authorizes putting the sleeping sentry before the firing squad apparently because he is not jitte ry enough even to stay awake.
There is also the statute’s strange relation with fear. A ll law must p a y homage to fear for if the law does not succeed in nurturing the passions that will make it self- enforcing, such as a sense of du ty or a special reverence for the law as law, it must have recourse to fear, the passion that underwrites all coercive law - fear of p u n ishment or the fear of the shame of being execrated as a law breaker. But this stat
ute places fear at its substantive core, for it is fear-impelled action that it m ostly seeks to regulate.
O n ly paragraph 6 - the stricture against looting - cares nothing about fear, not even the fear that you and yo u r raping and pillaging comrades inspire in the en
em ies’ civilian population as you quit yo u r proper place to plunder. Like the other provisions the anti-looting provision is devoted to maintaining the delicate bal
ance of forces that keep armies behaving as armies rather than as crowds. A t times that balance is as susceptible to being undone b y routing the enem y as by being routed by him. Success can be as disordering as failure4. The initial success of the German offensive on the western front in M arch 1918 was stopped, say some, as much by the German soldiers stumbling upon stores of wine and cognac as by A l lied resistance. But the weight of these strictures shows that loss of discipline and order bred b y greed, cruelty, lust, and other manifestations of exultant riot is of significantly less concern than the loss of discipline bred b y fear, slackness, and 3 See William B r a d f o r d H u i e , The Execution of Private Slo vik (N ew Y ork 1954).
4 O n crow ds and arm ies see J o h n K e e g a n ’s discussion in: The Face of Battle (N ew York 1976) 174-176.
Weak Legs: Misbehavior before the Enem y 17 failure of nerve. N a rro w self-interest in the exuberantly acquisitive style of the looter is just not as w orrisom e to an arm y as narrow self-interest in the life-pre
serving style of the coward. Fearfulness, not lust or gluttony, count as a soldier’s first sin.
There lurk in this strange statute various attempts at a theory of the moral and legal econom y of courage, cowardice, d u ty and fear in the context of the demands a polity, in this case the Am erican polity, makes upon its combat soldiers. The exposition that follows, structured m ostly as a gloss on the various provisions of the statute, seeks to reveal the features of that economy.
R unnin g A w ay
Isn’t running away, punished in paragraph 1, running like hell for the rear, pre
cisely how we visualize the purest cowardice (punished in paragraph 5), just as casting a w a y arms (punished in paragraph 4) so you could run a w a y faster was how Plato and Aristotle envisioned it5? In fact, the v ery vividness of the image of running a w a y has led some defendants to prefer being charged w ith the vaguer and more abstract cowardice under paragraph 5 considering it less prejudicial than an accusation of running a w a y 6. But statutory provisions that to the normal eye look duplicative w ill inspire interpreters to invent differentiating glosses, just as language itself, though needing all kinds of structural and particular reduncfancies, never quite allows a perfect syno n ym . So paragraph 5 - cowardice - was read to require a show ing of fear as a necessary element of the offense7. C o w ardice had to be motivated b y fear or it was not cowardice, but running away, it was decided, did not need to be so motivated. This strikes normal people, non-law yers, that is, as somewhat perverse. W h y else w o u ld anyone flee battle, run away, if not in panic, terror, or out of simpler fears of death and m ayhem ?
The m ilitary judges struggled to give running a w a y a meaning that w o u ld dis
tinguish it from cowardice. T h ey w anted to avoid defining running a w a y so ex
pansively as to undo the m ercy implicit in differently defined and lesser offenses such as “absent withou t leave”8, those acts of desertion that did not take place in the presence of the enemy. One m ilitary court became the final w ord on the sub
ject with this desperate attempt:
This term [runs aw ay] m ust connote some form of fleeing from an ensuing or im pending battle ... [Ijt appears that to lim it the phrase to flight from fear or cow ardice is too restricted.
3 See A r i s t o t l e ’s Ethics 5.2; 5.9. R hetoric 2.6; P la t o , L aw s, xii.944e. See also P o l y b i u s on capi
tal offenses in Rom an arm y, H istories 6.37-38, in: P o l y b i u s , The R ise of the Rom an Empire, trans. I a n S c o t t - K i l v e r t (H arm ondsw orth 1979).
4 See U nited States v G ross 17 (1968) U S C M A 610; 38 C M R 408.
7 U nited States v Sm ith (1953) 3 U S C M A 25, 11 C M R 25; U nited States v Brow n (1953) 3 U SC M A 98, 11 C M R 98; U n ited States v M cC orm ick (1953) 3 U S C M A 361, 12 C M R 117.
8 See 10 U SC S §885 (desertion); 10 U SC S §886 (A W O L).
