Wallace-Hadrill, A. - The Emperor and His Virtues. 1981

8/9/2019 Wallace-Hadrill, A. - The Emperor and His Virtues. 1981

http://slidepdf.com/reader/full/wallace-hadrill-a-the-emperor-and-his-virtues-1981 1/29

The Emperor and His VirtuesAuthor(s): Andrew Wallace-HadrillSource: Historia: Zeitschrift für Alte Geschichte, Vol. 30, No. 3 (3rd Qtr., 1981), pp. 298-323Published by: Franz Steiner VerlagStable URL: http://www.jstor.org/stable/4435768Accessed: 24/11/2010 07:19

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available athttp://www.jstor.org/page/info/about/policies/terms.jsp . JSTOR's Terms and Conditions of Use provides, in part, that unlessyou have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and youmay use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained athttp://www.jstor.org/action/showPublisher?publisherCode=fsv .

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printedpage of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

Franz Steiner Verlag is collaborating with JSTOR to digitize, preserve and extend access to Historia: Zeitschrift für Alte Geschichte.

http://www.jstor.org

http://www.jstor.org/action/showPublisher?publisherCode=fsv

http://www.jstor.org/stable/4435768?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp



http://www.jstor.org/page/info/about/policies/terms.jsp

http://www.jstor.org/stable/4435768?origin=JSTOR-pdf




THE EMPEROR AND HIS VIRTUES'

The power of the Roman emperor derived from many sources; from armedsupport, from legal and constitutional recognition, eventually also from thesheer inertia of a bureaucratic machine. Among other factors, a not negligiblerole was played by persuasion and belief. At least in part, the emperor waswhat Max Weber termed a 'charismatic' ruler; that is to say one whose powerdepends on the conviction of his subjects that he is personally in possession ofgifts or talents essential for their well-being, yet beyond the reach of theordinary mortal. He need not himself be regarded as a god (though indeed the

ambiguity of the emperor's position on the line that divides the divine and themortal is well attested by the manifestations of 'imperial cult'). The essentialrequirement for the charismatic ruler is the possession (in his subjects' eyes) ofpowers regarded as coming from outside, not from normal human nature.2

In discussing this aspect of the emperor, it is usual to turn to a body ofevidence relating to what are known as 'imperial virtues'. In the most

I An earlier version of this paper was read to the Oxford Philological Society on 24 May 1979. 1am grateful to members of the society for their comments. I have benefited greatly from criticismsof various drafts by Prof. P. A. Brunt, M. H. Crawford, S. R. F. Price and D. R. Walker. Anespecial debt is owed to Prof. C. J. Classen who allowed me to read and use an unpublished paperon the same subject. For views expressed and errors committed I claim sole responsibility.

Bibliography: the following are referred to by author's name and date of publication alone.

M. P. Charlesworth, 'The Virtues of a Roman Emperor: Propaganda and the Creation of Belief',Proc. Brit. Ac. 23, 1937, 105ff.

R. Frei-Stolba, 'Inoffizielle Kaisertitulaturen im 1. und 2. Jahrhundert n. Chr.' Mus. Helv. 26,1969, 18ff.

M. Grant, Roman Anniversary Issues (1950).M. Grant, Roman Imperial Money (1953).H. Kloft, Liberalitas Principis, Herkunft und Bedeutung. Studien zur Prinzipatsideologie (1970).

B. Lichocka, Justitia sur les monnaies imperiales romaines (1974).H. Markowski, 'De quattuor virtutibus Augusti in clipeo aureo ei dato inscriptis' Eos 37, 1936,109ff.

C. H. V. Sutherland, Coinage in Roman Imperial Policy 31 BC - AD 68 (1951).S. Weinstock, Divus Julius (1971).L. Wickert, 'Princeps', RE xxii, 2 (1954), 1998ff.

Note also the following abbreviations:BMC = Coins of the Roman Empire in the British Museum (1923 on). Strack i-iii = P. Strack,Untersuchungen zur romischen Reichspragung des zweiten Jahrhunderts, 3 vols (1931 - 1937).2 For Weber's views on bureaucracy and charisma see From Max Weber: Essays in Sociology,

ed. H. H. Gerth and C. Wright Mills (1947), 196ff. & 245ff.; Economy and Society, ed. G. Roth

and C. Wittich (1968), iii, 956ff. & 1t1 ff.; On Chanrsma and Institution Building, ed. S. N.Eisenstadt (1968) - all three covering much the same ground. Cf. now the suggestive observationsof P. Veyne, Le Pain et les Cirques (1976), ch. 4.



The Emperor and his Virtues 299

important recent study of such a 'virtue', liberalitas, ans Kloft concludes:

'The principate s, to use Max Weber's terms, to a considerable extent acharismatic orm of rule ... The concentrated effort of the panegyrists opraise the princeps as the incarnation of all virtues, and the correspondingpropaganda n coins and inscriptions serve this purpose'.3 The aim of thispaper s to raise some questions of principle about the status and function ofwhat are, so it seems o me, confusingly grouped ogether as 'imperial irtues'.

Modern discussion of the place of virtues in the ideology4 of the Romanempire moves from an old but still illuminating aper by Charlesworth 1937).His thesis was briefly this: For the mass of the population of the Empire he

legal and constitutional position of their ruler was an irrelevance. Whatmattered was their belief that he was right for them and that they needed him.This belief centered on his possession of certain virtues'. Spread belief n these,and the ruler's position was secure. The vehicle for propagation f belief wasthe official coinage, which frequently bore on its reverses he depictions ofvarious virtues' personified. The choice of the virtues so advertised ependedon assumptions about what the 'ideals' of a ruler were, which ultimatelyflowed from Greek philosophical hought about kingship, and which became'canonised' n the Golden Shield presented o Augustus.

This thesis was accepted with only minor reservations n the mostauthoritative tudy of the 'Herrscherideal', Wickert's Pauly article on thePrinceps, and by most subsequent cholars.5Yet there are certain eatures ofhis argument hat should have caused hesitation.6

1) He saw the propaganda of coinage as aimed at a wide social range(compared o the readership f the modern daily newspaper), ssentially hehumble as opposed to the educated elite ('the farmer n Gaul, the corn-shipperin Africa, the shopkeeper n Syria' p. 108). Yet the virtues propagated aresupposed to have been the product of Greek philosophy, the elite in its mostelitist of intellectual activities. What did the 'common man' care for Socraticvirtue ?

3 Kloft (1970), 181: 'Der Prinzipat ist, um mit den Begriffen Max Webers zu sprechen, zueinem betrachtlichen Teil charismatische Herrschaft . . . Das angestrengte Bemuhen der Panegy-riker, den princeps als Inkarnation aller Tugenden zu preisen, die entsprechende Propaganda aufMunzen und Inschriften, dienen diesem Zweck'.

4 Note that when J. Beranger Recherches sur l'aspect ideologique du pnonpat, (1953) discussesthe 'aspect ideologique' of the principate he is concerned with something a little different fromother, especially German, scholars. For him the 'ideology' is not a series of ideals for the ruler, butthe way in which the subjects perceive the function of their ruler (e. g. as one who undertakes agreat burden on their behalf). 'Virtues' consequently are of subsidiary importance in hispresentation.

s Wickert (1954), 2222ff. 'das Herrscherideal'; cf. 2231, 'Von einem Tugendkanon kann mannur mit Vorbehalt sprechen'. Kloft (1970), 181 n. 5; Lichocka (1974), 14; etc.

6 For attack on the idea of coins as propaganda, see below n. 49.



300 ANDREW WALLACE-HADRILL

2) He posited a 'canon' of imperial virtues, finalised by the Golden Shield:

virtus, clementia, ustitia, and pietas. Now though he admitted that there wereothers (he stressed providentia), still these four great qualities... .were alwaysthought of a present in and exercised by his [Augustus'] successors, and theywere certainly cardinal virtues of a ruler' (p. 114). Wickert lists some fifty andmore qualities attributed to emperors over the centuries by various sources,literary, epigraphic, numismatic; nevertheless, he only discusses the 'cardinal'four. But the very notion of a 'canon' ought to have been questioned. A canonimplies a fixed and generally accepted belief in a set of entities. It thereforeallows no room for difference of viewpoint between different groups of people

(let alone a distinction between 6lite and masses). Nor does it allow fordifference between different periods and historical circumstances, even be-tween different societies. Did the philosopher in Ptolemaic Alexandria, thehistorian in Trajanic Rome, and the panegyrist in late Roman Gaul really eachsee the same set of ideals in his ruler? If so, one must suspect that the idealswere so general and superficial as to be without any real value.

In view of these difficulties it is necessary to examine the evidence again. Ishall look first at the arguments for a 'canon of virtues', then at the relationshipbetween the 'virtues' met on the coinage and those in the literature of the

educated elite.

1. The Canon of Virtues

At some stage in 27 (or 26) BC the senate presented the newly namedAugustus with a Golden Shield, 'virtutis clementiae iustitiae pietatis erga deospatriamque caussa'. His pride in the honour is evidenced by the record in theRes Gestae, by the frequent representation of the shield on the coinage, and bythe dissemination of copies throughout the empire, of which that from Arles

survives asa

specimen (Pl.1,1). But even

settingaside the

assumptionthat

Augustus designed the wording himself as a 'political platform',8 the canonicalstatus of these virtues is a delusion.

The first premise on which the argument rests is that the virtues of theshield, Virtus, Clementia, lustitia and Pietas, are identical with those of Greekphilosophy. The identification is widely held to have been proved by thePolish scholar Markowski (1936). Yet it is either gravely misleading or simplyfalse. The facts are straightforward enough: no doubt that moral philosophers

7 On the circumstances of the presentation of the shield see W. K. Lacey,JRS64, 1974, 181-2,arguing that the date Cos VIII of the Arles shield may be correct. For a collection of evidence ofrepresentations of the shield, and discussion of the significance of the virtues, Tonio Holscher,Victoria Romana (1967), 102-112. See also H. W. Benario, ANRW II, 2 (1975), 80ff.

8 So explicitly I. S. Ryberg, 'Clipeus Virtutis', in The Classical Tradition Studies in Honor ofA. Caplan) ed. L. Wallach (1966), 233. Cf. Charlesworth (1937), 112.




constantly spoke of four cardinal virtues, but they are not the same ones.Bravery (&vbQsa(), Temperance (owqpeoou'v), Justice (8&xaLoOmvvq) ndWisdom ((fQovroLg/Go0Pca) form the canon. The group is of course Socratic inorigin. The earlier dialogues of Plato treat of five parts of &QDEx, these four andoOLT1T; Fioa?EicEa),9 nd this group of five appears in the classic encomium ofa king, the Agesilaos of Xenophon (a Socratic).'0 But religious observance is oflimited appeal to a moral philosopher, and from the Euthyphro on Plato dropsOCLOLTJ;, and limits the canonical number to four." The analytic Aristotledrops this restriction and extends the range (without reintroducing oo-Lm;).'2It is the Stoics who confirm the idea of a canon, reestablishing the Platonicfour, and turning other virtues into subspecies of these.'3 By the time of Cicerothis is firmly established, and he constantly rehearses the four as an assumedfact of moral philosophy: he renders them as fortitudo (not virtus), temperan-tia/continentia, iustitia, and prudentia/sapientia. What canonised this groupperhaps more effectively than Stoic systematisation were the rigid prescrip-tions of rhetoric: in rhetorical handbooks, whether Cicero and Quintilian, orthe arid imperial tracts collected in Spengel's Rhetores, the prescription isconstant, that to praise a man, in particular a king, the orator must demonstratethe four virtues."4 This was the group which passed via late antiquity to the

middle ages. A Carolingian Gospel illustration showing a king surrounded bythe classic virtues of Prudence, Justice, Temperance and Fortitude may betaken as representative of the rich postclassical tradition (PI. 1, 2).'5

' Traces of the &Qrtai as a canon before Plato are dubious. Gorgias Epitaphios 82B6 Diels-Kranz mentions Q?V . . . IO nQ&OV trLELXtg . . . ;XCLLOL . . . Vb1OEdLg, but not in such a way asto suggest these are the main parts of virtue. For Plato's division of &QE"T nto 5 1IoQLa e. g. Prot.349B. See also 0. Kunsemuller, Die Herkunft der platonischen Kardinaltugenden Erlangen1935 = New York 1979).

1 Xen. Ages. iii (eVtiOIOELa), v (bLxaLou6vvl), v (owWQoolvrl), vi, 1-3 (&896vQa); vi, 4-8

(oo(p(a). Markowski p. 121 misleadingly omits mention of oo(pLa.1 Rep. v, 428A T?TTaQa' Ovtc uyxdvEt. A. Dihle, DerKanon derzwei Tugenden(Koln 968),15ff. describes Plato's establishment of a 'canon' on the basis of 'Vulgarethik', and the' eventualelemination of 6Lot6ri;, which survives in vulgar ethics in the pair 6uo; xai btxatog. For a sketchof the history of the Platonic canon, H. North, 'Canons and hierarchies of the cardinal virtues inGreek and Latin literature', The Classical Tradition ed. L. Wallach 1966), 165ff.

12 For &QE-taCn Aristotle EN iii-vi; cf. EE iii, Magn. Mor. i, 20-34, de Virt. 2, 4 & 5.Markowski p. 112 asserts that tQa6rTrl; eplaces wisdom as the fourth cardinal virtue; but it is onlyone of several non-Platonic virtues of Eth. Nic. iv, and (Pe6otL; is fundamental to Aristotle'sscheme. For a brief list cf. Rhet. 1366B1ff.

'3 SVF ii, 262ff. for constant repetitions of 4vb&eca, OWq(oo0v0V (tyxQD&eLa), bLxatooOvv,

(qpLvt6otg, lready from Zeno (i, 47).14 See J. Martin, Antike Rhetorik 1974), 177ff. For the debt to philosophy, W. Kroll, Philol.

40, 1935, 206ff.15 See the careful study of Sibylle Mahl, Quadriga Virtutum. Die Kardinaltugenden in der

Geistesgeschichte der Karolingerzeit (1969), esp. 171-6; and in general Helen North, From Myth toIcon: reflections f Greek ethical doctrine n literature nd art (1979), 198ff.




The Golden Shield tells a different story. Virtus, though close to etvb ELa, isnot used to translate it."6 Clementia is certainly not the same as oCopQooVi,and it involves an element which the Stoics were notorious for rejecting,forgiveness."7 There are indeed passages where Cicero makes clementia a sub-category of temperantia; but in other passages it is subsumed under othervirtues."8 Justice is canonical. But, most important discrepancy of all, there isno confusing pietas and sapientia.9 If one of the original five virtues was to bedropped, a philosopher would never be prevailed upon to abandon wisdom;whereas piety did indeed lapse.

It is surely clear that the most one can argue is that the Golden Shieldrepresents a variation upon the canon. In this case we must begin by askingwhether variations are in fact found? After all, it goes against the nature of acanon to admit variation. Here one comes up against a problem of method.Groups of four virtues may be mentioned often enough, but how are thosewhich are deliberate variations on the canon to be identified? The pitfall isarbitrary selection (as when Weinstock takes a group out of the middle of along list of virtues and identifies them as canonical).20 n practice, I have foundthat the context occasionally provides hints when a variation is intended: mostclearly when four virtues are used as the basis of divisio of a passage, or when

enumerated in the form 'a, b, c, d and all the other virtues'.In a philosophical context, minor variations are possible. The first book of

de officiis is based on the four virtues of the good citizen: among thesemagnitudo animi replaces fortitudo, but this, as Cicero explains, is becausemagnanimity also includes patientia, passive as well as active bravery, andshould therefore be regarded as the leading species.2' In a rhetorical contextCicero uses Aequitas in the place of Iustitia: this and the three others areranged on his side against the Iniquitas and other vices of Catiline. 'Iniquity'

16

Cicero, though his use of virtus is outstandingly frequent, never uses it to translate dev6Q6ain the canon; in general cf. W. Liebers, Virtus bei Cicero, Diss. Leipzig 1942; W. Eisenhut, VirtusRomana 1973 and RE Suppl. xiv, 1974, 896ff.

1 See recently M. Griffin, Seneca : a Philosopher in Politics (Oxford 1976), 155ff.18 For clementia s a part of temperantia ee von Premerstein, Vom Werden M. Wesendes

Prinzipats, 8f.; Helen North, Sophrosyne 1966), 300f. But though at Cic. de Inv. ii, 164 clementiais treated thus, at de off. i, 88 it is handled under magnitudo animi, and at Part. Or. 78 lenitas npunishment is a sub-division of iustitia. Similarly Menander Rhetor iii, 374, 28f. and Aristides ix,16-24 treat cptkavfgwnLat under bLxaLootuio.

19 pietas is absorbed under the heading of iustitia rather than sapientia) by Plato Euthyph. 12E,Cic. Part. Or. 78.

20 Weinstock (1971), 228 quotes de Or. ii, 343 for the combination of 'clementia, iustitia,benignitas, fides, fortitudo'. The context makes it quite clear that he has not abandoned his canonof ii, 45f. The principle of arbitrary selection is fundamental to Markowski's argument (see nn. 9,10, 12).

21 de off. i, 61-92; cf. Part. Or. 77. For <ayaXoVuXcia s a Stoic subdivision of &vbQe(a andCicero's inversion see U. Knoche, Magnitudo Animi (Philol. Suppl. 27, 3, 1935), 51.




sounds better than 'injustice', perhaps.22 The same variation is used in a

panegyric on Julian: the orator sees in him the very form of Virtue - thecandour of Equity, the blush of Temperance, he stiff-neck of Fortitude andthe penetrating lance of Providence. The hearers will quickly understand hatAequitas is bLxatLOcnUv, and Providentia is pQovrnoLg.23

Another type of variation is when the context provides the reason why oneof the canonical four has to be dropped and replaced. Cicero explains why thelawyer Sulpicius could not reach the consulship: he has the virtues of'continentiae gravitatis iustitiae fidei' which qualify him for the consulship.But he lacks prowess in war; and what is the use of legal learning in the

consular elections? Fortitudo and sapientia are therefore dropped (pro Mur.23). Or Pliny praises a friend as being the equal of philosophers in 'castitatepietate iustitia fortitudine'. He could hardly pretend he was their equal insapientia, so pietas s resuscitated for the context.24 A panegyrist introduces anold debate, whether Virtue or Fortune won Theodosius his battles? Con-stancy, Patience, Prudence and Fortitude all bear witness on the emperor's side- two of them canonical virtues, two of them replacements for the lessappropriate Justice and Temperance.25

Finally, there are cases where there is little more than the number of four

that suggests the author wishes to allude to the philosophical canon. WhenCicero requested a triumph for his work in Cilicia, Cato opposed the request:instead that irritating Stoic voted him, what he least needed, a Certificate of hisintegrity, justice, clemency and fides (ad Att. vii,2,2). Were Cato not sophilosophical, one would hardly guess there was any significance in thenumber of four.26

Where does this leave the Golden Shield? On balance, I am inclined to thinkthat, as with Cato's testimonial, the number of four is indeed meant to give theappearance of the philosopher's canon. But what dictated the choice of the

actual virtues mentioned? Given that only one of the quartet, lustitia, is

22 Cic. in Cat. ii, 25; but cf. de Or. i, 56; defin. ii, 83; Livy iv, 6, 12 for the substitution ofaequitas for iustitia.

23 Pan. Lat. iii (xi) 5, 4. For providentia as a subdivision of prudentia, Cic. de Inv. ii, 160.Similarly CIL vi, 1741, 3f. celebrates Memmius Vitrasius Orfitus as distinguished 'ad exemplumveterum continentia iustitia constantia providentia omnibusque virtutibus'. Constantia represents&6V6QEa.

24 ep. i, 22, 7; cf. iii, 2, 2 where Arrianus Maturus is praised as 'princeps ... castitate, iustitia,gravitate, prudentia'. For gravitas as a variant see below.

25Pan. Lat. ii (xii), 40, 3. For other variants in the Panegyrici see viii (v) 19, 3 'gravitas, lenitas,verecundia, iustitia', cf. vi (vii) 4, 4; and ix (iv) 8, 2, 'continentia, modestia, vigilantia, patientia' as

the desirable products of rhetorical education.26 Cicero himself in praising Pompey, though employing the fourfold disposition of rhetoric,

chooses quite different heads: scientia rei militaris, virtus, auctoritas, felicitas (Imp. Pomp. 28).Three of these are echoed by Ammianus on Julian (xxv, 4,lff.).




unquestionably 'philosophical', and given the wide range of variations shownto be possible by those who did abandon the exact canon of the Stoics, it seemsmisleading - or pointless - to assert that Augustus picked them because theywere known to be the four virtues of the ideal ruler. In fact, the philosophicalvirtues were acknowledged as being partly inappropriate to a ruler. Defendingthe king Deiotarus, Cicero mentions his frugalitas: his is the greatest virtue,he says, modestia, r temperantia; ut it is no way to praise a king. 'Brave, just,severe, grave, magnanimous, generous, beneficent, liberal - these are regalcompliments (hae sunt regiae laudes); the other one is for private citizens'(Deiot. 26).

Reminiscent perhaps, then, but not identical. But all this will be beside thepoint if the quartet of the Golden Shield in fact acquires canonical statusthereafter. How are we to judge? One expects of a canon, whether of sevenSages, nine Lyric poets, or four Virtues that it should be repeated, preferablyad nauseam, r at least with a tolerable frequency.27 It is a curiosity, then, thatnowhere in the literature, at least as known to me, do Virtus, Clementia,lustitia, and Pietas occur together in a context that suggests their special status.Certainly they are among the most frequently named individual virtues in thelate imperial panegyrics; but never together as a group.28 The imperial coinageis of course the direction in which Charlesworth indicated we should turn: butonly Hadrian, Pius and Marcus, of so many dozen Caesars, do actually minttypes of all four virtues (see Appendix with chart). Even here there is no specialassociation. The types issued by imperial mints varied from year to year, andnone of these three emperors ever issued all four types from the same mint in

the same year.29 t is a story of missed opportunities. Right at the start Tiberiuscame close to achieving the Four. But issuing a series of handsome femaleheads (probably Livia) he labelled them as lustitia, Pietas and - Salus. (Pl. 2,1-3).3?. In another emission he came up with Clementia, inscribed round a

commemorative shield: but twinned it with Moderatio (PI. 2, 4-5).3 Even

27 Cf. Radermacher RE x, 1873ff., s. v. Kanon. But see R. Pfeiffer, History of ClassicalScholarship i (1968), 207 against the mistaken conception that 'canon' in this sense is an ancientterm: it stems from the 18th century.

28 The closest approach is Pan. Lat. x (ii), 3-4 where all 4 are mentioned. But they are notlinked. The emperor is in peace a model of iustitia and virtus (3,3); his conduct of wardemonstrated not only fortitudo, but clementia and pietas (4, 3-4).

29 The reason lies in the great rarity of both Clementia and lustitia as types; only Tiberius,

Hadrian, Pius and Marcus have both, and none of them produce these types except in isolatedemissions. Only in the series of AD 128 (below) do both occur in the same series, and here Pietasand Virtus are both missing. For details see Appendix with chart.

30 BMC i, 131 & 133. For doubt whether all three represent Livia, Grant (1950), 37, Sutherland(1951), 96; Lichocka (1974), 24f.

31 See Appendix.




when under Hadrian, as will be seen, the first conscious series of virtues wasminted, Pietas and Virtus were omitted.

To this argument silentio one may add a positive one. If the Golden Shieldwas regarded as an authoritative variation on the philosophical deal of astatesman t ought to have been exploited in contexts where the ideal wasdrawn upon. This is where the rhetorical tradition is significant. Thehandbooks continued to recommend he traditional Platonic quartet for thefakZLXLX;X6yog. Menander, he best of them, acknowledges qPLXavfQonTa(clementia), but treats t as a sub-division of 6LxaLOoluvq, not as an alternativeto aowqonUvrj. What is recommended n theory is put into practice by the

anonymous panegyric that has come down under the name of Aristides.32Again in practice he late imperial Panegyrici are well aware of the theory ofcardinal virtues. But despite at least three occurrences f the classical canon,and five variations on it, they never hit on the Augustan group.33

In view of the pattern of the rhetorical radition, here seems to me littlepoint in constructing hypotheses that detect the Golden Shield virtues inobscurer corners of the sources. Horace's Roman Odes are one old huntingground,34 ertain sarcophagi (for private citizens) of the Antonine periodanother.35 ut without solid foundations, uch hypotheses must totter. More

weighty is Stefan Weinstock's attempt to trace an anticipation of the'statesman's virtues' by Caesar. But he freely admits that the evidence foreither iustitia or pietas playing a significant part in Julius' propaganda snegligible; and all in the end boils down to a pamphlet detected behind theaccount of Romulus in Dionysius' Roman Antiquities.36 ven granted thehypothesis that his source was a pamphlet and its date Caesarian, he case, asBalsdon saw, will not stand.3' Dionysius talks not of the qualities of thestatesman, but of the qualities Romulus nstilled nto the Romans. These arepiety, temperance, ustice and nobility in war. While it is just arguable hat

clementia may substitute for o(oppocnJvri n a canon, it seems ludicrous toimagine hat a Roman reading how Romulus brought w(Joxpocnuv o his stateby legislation ontrolling he lasciviousness f women would be put in mind ofthe clementia Caesaris.

After making so many negative points about the Golden Shield, I oughtperhaps o add something positive. The context to which the Shield urely does

32 For refs. above n. 18.33 Variants are cited above. The canon at xi (iii) 19,2; vii (vi), 3,4; iii (xi), 21,4.31 Von Premerstein, Vom Werden nd Wesen of.The arguments of I. S. Ryberg op. cit. (n. 8)

detecting the 'canon' in the Roman Odes, the Aeneid, Ovid Fasti ii, 140ff. and the Ara Pacis, donot bear detailed examination.

35 G. Rodenwald, 'Ober den Stilwandel in der antoninischen Kunst', Abh. Ak. Berlin 3, 1935,6f.

36 Weinstock (1971), 243 & 248 admits the lack of evidence.37 Balsdon, JRS 61, 1971, 22f. The relevant part of the 'pamphlet' is Ant. Rom. ii, 18ff.

20



306 ANDREWWALLACE-HADRILL

belong is indeed a Greek one, but not a philosophical one. One has only to

open Dittenberger OGIS almost at a random page to discover that it wasstandard practice or the Greek cities in the Hellenistic period o honour kingsand other benefactors by presenting hem with crowns, statues and the like,bearing honorific citations.38 o a city presents Attalus III of Pergamum withan agalma and eikon; they are to bear the respective itations of, "O 6[IioqkatUkXa ATTakov ... &QFCij KVEXEV xaci &v6Qaya*ia;g Tii XCLTCLt6EILOV

.' and "O bo;g Paaolkta Attakov.. . a&ETTI EVEXEV xCii( PvrGEg ...xai pWyaXopE[EiQ;ag fi ds 'Eavtov' [i. e. to the Demos].39 There are manyvariations, but the commonest features are the start in the form &tgujg 'vExEv

and the conclusion rvo(a;/E EQye0Lag/EU 43EWLa ; ;LcEaut6v or 'both tothe gods and to the city'. Romans knew this custom well, because they asgovernors inherited the honours (numerous examples in IGRR). In the case ofCaesar, a whole rash of such honours breaks out after Pharsalus (Raubitschekcollected the different versions).40

The precise wording of the citation on the Golden Shield is not preserved byany single source; but by conflating them the following may be achieved:'Senatus Populusque Romanus Imp Caesari Divi F. Augusto Cos VII deditclipeum virtutis clementiae iustitiae pietatis erga deos patriamque caussa'." If

we compare the honorific decrees of the Greek tradition, the following pointsare found to be paralleled.i). The form of the honour: though a golden crown was the traditionaldecoration, golden shields are regular variants in the last century BC, as theone presented to Q. Cicero by his province.42ii). The form of citation: 'The people honours so-and-so on account of...'iii). The characteristic initial 'on account of virtue and ... ':&QeTis 9VEXEV'virtutis causa'.iv). The common 'both ... and ...' flourish at the end, 'pietatis erga deos

patriamque'. So exactly a dedication to Caesar at Pergamum: [FoE L'CLtItQ6;] E[] To[iv51O]Eo[(i ilv T]E o6[ktv] (IGRR iv, 306).

38 This goes back to the honorific practice of the Athenian assembly. For a convenientsummary see Larfeld, Handbuch der griechischen pigraphik i, 836f. (for Athens); i, 509ff. (forthe rest of Greece).

39 OGIS i, 332. For similar passages see the inscriptions cited by W. Schubart, Archiv urPapyrusforschung 2, 1937, 5.

40 Raubitschek, JRS 44, 1954, 65ff. Add IGRR iv, 306.41 This is based on the Aries copy which preserves the formalities. Cos VIII is changed to Cos

VII but may in fact be correct (see Lacey JRS 1974, 181) and the missing (caussa) is supplied fromRG 34. No confidence can be placed in the connectives between the various virtues found in theRG version, which has abandoned the formal citation for an indirect report. (I thus implicitlyreject all speculation based on these connectives from Markowski on.)

42 Macrobius Sat. ii, 3,4 for Q. Cicero. The form of honour is normal in the last century BC cf.OGIS 571 n. 4; Sutherland, JRS 49, 1959, 137 n. 28. Shields usually bore an image of thehonorand; for one with an inscription only, OGIS 767,26 (Augustan, from Cyrene).



The Emperor nd his Virtues 307

v). The intervening iustitia is matched by btxatoo(vr in many Greekcitations, ncluding wo to Caesar IGRR iv, 305 and IG viii, 1835).43

Only two points about Augustus' shield are unusual: the mention ofclementia neither tELUXEMa nor (Ltkav*QOXTLa ccur in similar contexts) andthe selection of a group of four virtues two or three are usual, hough four dooccur in a dedication rom Lindos of 41 BC, OGIS 765, 60). If the secondpoint suggests that the senate was trying to superimpose he impression of aphilosophical anon, the first does not. Both points, however, are paralleled nCato's Testimonial o Cicero mentioned above.

Augustus' Golden Shield, then, is perhaps better regarded s the end of an

old tradition, rather than the beginning of a new one. Virtus, Clementia,lustitia and Pietas individually may all have been important, ndeed central'virtues' n certain contexts under the empire;' but as it seems to me, there ssimply no evidence or supposing hat the Shield played any authoritative olein spreading belief in them, nor that they constituted a 'canon'.45

2. 'Virtues' nd the coinage

If the evidence for a 'canon' of virtues, whether officially propagated or

spontaneously adopted, breaks down, we must be led to question further.How far was there an official attempt o propagate elief n any sort of virtues?Especially f we assume for the purposes of argument) hat propaganda wasdirected at the 'masses', what sort of appeal would 'virtues' have had, and ifthey had any at all, what kind of virtues are stressed?

Charlesworth's ase rested on the imperial oinage, as it must. Inscriptionsmay provide valuable upplementary vidence; but they cannot offer anythingas a corpus on the scaleof the coinage, regularly roduced, ully preserved, ndsystematically atalogued by modern scholars. Note however one important

deficiency of imperial catalogues, that they take no cognisance of localcoinages.46) he thirties was an ideal decade or detecting mperial propaganda:

4 XCtLoOhrv- s fairly common in citations: see conveniently Dittenberger, Sylloge3 IV, 326,Index s. v. EVEXa.

" Above all it should be borne in mind that the context of the shield is victory in civil war: it isnormally represented on the coinage as carried by a flying Victory. So (in my view) rightly R.Combes, Imperator (1966), 438f. This helps to explain the preeminence of these virtues inConstantinian panegyrics, that revolve constantly around victories in civil war.

45 A further argument has been employed from the celebration of the clupeus tself on thecoinage: Strack i, 57-61; cf. K. Kraft, Jahrb. Num.

Geldgesch.2, 1962, 7ff. But though the shield

itself was remembered it does not follow that the virtues it celebrated also were. Telling against thishypothesis is the fact that the accompanying legend of CL(ipeus) V(irtutis) is replaced under Neroby VICT(oria) AUG(usti). What was remembered was the victory not the virtues. CL V is onlyrepeated in a direct revival of the type in the civil wars, BMC i, 304f.

' See A. Burnett, JRS 68, 1978, 173ff.

20*




not only was propaganda itself very much a live issue,47 but the numismaticmaterial was for the first time properly catalogued, indexed and discussed withparticular awareness of its 'ideological' content, by Mattingly in RIC andBMC; while simultaneously in Germany Paul Strack was working over theAntonine Reichspragung with meticulous care.

Numismatic tastes are rather different these days; and though as late as 1959Sutherland defended the notion that coin-types were both understandable andunderstood, most of his colleagues shifted onto pastures new and economic.48The conception of coins as an organ of propaganda now meets with increasingscepticism.49 Too little is known of the mechanics of type selection (how farwas it by the emperor, how far by his subordinate personnel?). Nevertheless,the types are there, in a variety exceptional in numismatic history, and theymust in some sense reflect 'official' perceptions of the emperor.50 Which'virtues', then, are advertised, and why? Can we speak of an attempt by thosein authority to persuade the subjects that their ruler was the right man for thejob?

Working from the basis of a crude chart of the personifications on Romancoinage (see Appendix), I wish to make three observations. The chart does not,it is important to remember, cover all the themes of imperial coinage, specific

events like Aegypto Capta or Quadragesima Remissa, scenes like a PraetorianDecursio, the new harbour of Ostia or the crowning of emperor by senate. It isrestricted to what is relevant to the question of virtues, the (predominantly)female personifications.

I) Personifications and virtues

The first point is about the use of the label 'virtue'. The category we arediscussing is an ancient, not a modern one (&QEt-af, irtutes), and it is as well tobe clear what was meant by it. There is no problem here, as philosophersdefine it often enough. As for Aristotle &QnM s a 9l; wvXrg, so for Cicero'virtus est animi habitus' (de Inv. ii, 53) or an 'adfectio animi constansconveniensque, laudabiles efficiens eos in quibus est' (Tusc. iv, 34). Virtue is

47 Charlesworth cites in his bibliography L. W. Doob, Propaganda. Its Psychology andTechnique 1935). The author had recently travelled in Germany, and has interesting observationson the Nazi propaganda machine.

48 JRS 49, 1959, 46ff., against A. H. M. Jones, 'Numismatics and History', in Essayspresentedto Harold Mattingly (1956), 13ff. = The Roman Economy, ed. P. A. Brunt (1974), 61ff. M. H.Crawford's note at the end of the reprint (80f.) shows how influential Jones' article had been.

'9 For scepticism among numismatists about 'propaganda' see Belloni, 'Significato storico-politico delle figurazioni e delle scritte delle monete da Augusto a Traiano', ANRW 11, 1 (1974),997ff.; T. V. Buttrey, 'Vespasian as Moneyer', NC 1972, 89ff.

50 Cf. S. Price, CR 29, 1979, 278f. citing several texts that suggest that the emperor was heldresponsible for coin types, especially the anonymous de rebus belicis 3,4. Also Sutherland, JRS1959, 52 for speculation on officials in charge of coin-types.




the moral quality of a man, whether innate or developed by education and

practice. The trite definition of rhetoricians s also relevant. There are threepoints for which a man can be praised: he things of the soul, the things of thebody, and external hings (like wealth and luck). Only the first, t'a tiqg Vx~i;,constitute proper 4QETT.5

But the usage of modern numismatists s different, and varied. Mattingly52started rom a passage of Cicero (de leg. ii, 28) that distinguishes irtutes romres expetendae, blessings to be sought; but having acknowledged hat manypersonifications were of desirable tates rather han virtues, he proposed tokeep the term 'virtutes' or all. In a way he was right. The Tiberian eries of

Justitia, Pietas and Salus (Pl. 2, 1-3) are so clearly inked n conception hat itmakes no numismatic ense to distinguish hem as different ypes of things,although Salvation/Safety s no EML; VUXii.5"Moreover, there is a sense inwhich (e. g.) Salus s more than a res expetenda; ne of the regular eatures ofthese personifications s the attachment f an explanatory Augusta or Augustito the legend. The implication s that the quality, like a virtue, resides withinthe emperor so is not one of tat E'.)f?v): Salus Augusti s not just the Safetyof the ruler, but the Saving Power that flows from him.54

Moving from this last observation, Michael Grant attempted a fresh

distinction: the adjectival Augusta signified a res expetenda, a blessing, thegenitive Augusti a virtue, a quality within the ruler.55 his was a distinct stepbackwards. He lists sixteen personifications ualified n the 1st century AD byAugusti. Among these are perhaps ive which Cicero would have called virtues(Aequitas, Clementia, Constantia, Pietas and Virtus); others are, philosophi-cally speaking, ta Kw&,v - Victoria, Tutela, Libertas, Pax, Securitas,Aeternitas, Fortuna). Now doubtless these all reside within the metaphorical'godhead' of the emperor: but what sort of theological sophistry is needed toturn Annona Augusti into a virtue?

My point, then, is this. Numismatic specialists are entitled to use whateverlabel they choose for these personifications. But if one is to compare coins withother sources, particularly philosophically inspired ones (i. e. in talking of thevirtues of the ideal statesman) it is vital to distinguish what is a virtue and whatis not. Among the forty or so personifications of the imperial coinage, only a

51 See Martin loc. cit. (n. 14).52 'The Roman "Virtues"', HThR 1937, 103ff., an important statement of his position.53 It is perhaps better to follow Wissowa, Religion und Kultus 328 in saying there is no

'Wesensunterschied' between blessings and virtues because both are seen as gifts of the gods - notas human dispositions.

5 Cf. BMC i, lxxiv & ii, xxif. on Salus.55 Grant (1953), 154ff., esp. 167. The distinction of Augusti and Augusta is pressed by Strack i,

49-52; yet the frequency of the ambiguous abbreviation AUG pleads against precision. Cf. K.Latte, Romische Religionsgeschichte (1960), 324 n. 1.




dozen are virtues.56 t is also worth noting that the types which only appear

once or twice are almost all virtues.57

II) Pattern of distribution

The second point is about the pattern of distribution of personifications.Charlesworth draws inferences rom the presence or absence of individualvirtues in the coinage of individual rulers - one of his most widely quotedobservations s that clementia was not advertised by most early emperors,which he explained by the 'despotic character' f the virtue.58 ut this kind ofobservation can only hold good if the general pattern would lead oneotherwise o expect clementia. If, of course, he coinage acted as a continuationof the message of the Golden Shield,one would expect Clementia. But does it?

An overall conspectus of the personifications eveals a pattern which thespecialists have never, to my knowledge, made explicit. Three periods may bedistinguished. In the first, the Julio-Claudian period, personifications rescarce and spasmodic. Such as do occur are not repeated rom reign o reign.There is no sign of a systematic attempt o put across any message, et alonethat the Shield virtues have been realised. The intermittent appearance fclementia is no surprise: pietas and the others are no less intermittent. Giventhe internal history of Roman coinage all this is natural nough: t took time todiscover the value of these personifications as reverse types. Slogans withpersonifications sually in the form of female heads, Libertas, Pietas, Felicitasand Virtus etc., had appeared nder he late republic cf. P1.2, 8-11). Tiberiustentatively revived this tradition (ignored, significantly, by Augustus.) It isworth noting that one of the initial attractions f this type of reverse was that tprovided an excuse for depicting female members of the imperial household(cf. Livia as Salus, Pl. 2,3). Caligula's hree sisters appear with the attributes fSecuritas, Fortuna and Concordia (Pl. 2, 6): they are among the first

'goddesses' o be depicted full-figure on imperial coins, which subsequentlybecomes the standard method of depicting personifications.5

56 I. e. Aequitas, Clementia, Constantia, Indulgentia, lustitia, Liberalitas, Munificentia,Patientia, Pietas, Providentia, Pudicitia, Virtus. I exclude Moderatio, Magnificentia, (andDisciplina) which are never personified. On Tranquillitas cf. below Appendix. The total of 40 isthe number of personifications listed by Gnecchi (see Appendix).

5 I. e. Constantia, Magnificentia, Moderatio, Patientia. Equally rare is Disciplina, not a truevirtue.

58 Charlesworth 113; Sutherland JRS 28, 1938, 129ff. corrected him on an omission in thenumismatic evidence, but his suggestion that clementia was 'too despotic' is still quoted withapproval, e. g. Wickert 2243. Yet if the virtue was not 'too despotic' for Seneca, it was hardly so forthe coinage.

59 BMC i, 152; Sutherland (1951), 152. Earlier is the unidentifiable personification of the seriesstarting in AD 13, BMC i, 124f., cf. Sutherland 84.



The Emperor nd his Virtues 311

The second period stretches from the civil wars of 68/9 to Antoninus Pius.This is the heyday of the personification. Not only are the goddesses found inunprecedented number (32 as against 16 in the first period); a quite newpattern has emerged of repetition and continuation. Once one emperor hasintroduced a new type, it is notable when his successors do not continue it.There can be no doubt what gave the impetus to this new pattern.w In the warsof 68/69 the contending parties, starting with Galba and Vindex, used thecoinage, necessarily minted to finance their war-effort, to advertise their hopesand ideals (e.g. Bonus Eventus, Concordia, Libertas, Virtus: PI. 2, 12-15).Appropriately enough all their personifications (except Securitas) have repub-lican precedents (cf. PI. 2, 8-11). The ideals of the insurgents are taken up byGalba as emperor (note that by now all the republican personifications havebeen revived); Galba's are continued by his rivals, until with the Flavians thispattern of repetition settles down to become the norm (cf. PI. 2, 16-19). Butnew themes are constantly added, until with Hadrian and Pius comes theclimax.

The last period, from Marcus onwards (to, say, Diocletian) is onlydistinguished by its dullness. The repetition of types continues, more and moremeaninglessly. It is most seldom that a new type appears, and those arevariations on old themes (Perpetuitas for Aeternitas, Abundantia for Annona).It is as if the mint was rehearsing a doxology of empire established by thecentury that culminates with Pius. It is no coincidence that the loss of interestby numismatists in 'imperial virtues' coincides with the time the BMC movedinto this latter phase.6'

III) Hadrian and Virtues

The last observation puts together the results of the first two. Virtues properare a relative rarity among personifications; and it is only after 68 AD that anypersonifications, let alone virtues, acquire regularity. Can one speak of anysystematic attempt to propagate belief in the virtues of the ruler? To beginwith, claims to virtue are isolated and idiosyncratic: thus Tiberius has hisClementia and Moderatio (Pl. 2, 4-5), Claudius his Constantia (PI. 2, 7). Whenregularity supervenes, virtues are in a small minority: Virtus from Nero on,and Pietas more and more frequently. Aequitas becomes a regular from Galbaon: but there is a special reason, unconnected with the moral qualities of theruler. Aequitas refers almost certainly to the proper operation of the mint, and

60 R. H. Martin, Die anonymen Mu?nzen desJahres 68 n. Chr. (1976) now argues that the issuestraditionally assigned to Vindex, Galba and the Rhine armies were all minted by Galba in Spain.

61 Note the comment of R. A. G. Carson in the introduction to BMC vi (Severus Alexander -Balbinus and Pupienus 1962), 29, dismissing the reverses as little more than the 'ringing of thechanges on conventional and banal types'.




the legend is interchangeable ver the centuries with that of Moneta.62 t is

only after Nerva that things begin to look up: he establishes regular ype ofJustitia, Trajan one of Providentia. But Hadrian is outstanding: all hisinnovations are virtues Liberalitas, ndulgentia, Patientia, Pudicitia, Tranquil-litas) or quasi-virtues Hilaritas and Disciplina).63The case of Liberalitasemphasises how far this represents a new way of thinking about coin types.From the reign of Nero individual congiaria re celebrated y emissions withvivid depictions of the scene of the dole. But under Hadrian (tentativelyanticipated y Trajan) he abstract notion of Liberality cquires redominanceover the concrete scene. Either he goddess replaces one of the attendants, nd

the legend becomes LIBERALITAS UG, or the goddess alone represents hewhole scene by a sort of pictorial shorthand. This is typical of a general hiftfrom the specific, characteristic f Julio-Claudian everses, o the abstract.64What is most remarkable s the series issued by Hadrian n AD 128,which isresponsible or most of the new 'virtues'. In this year appeared n parallelClementia, Indulgentia, ustitia, Liberalitas, Patientia and Tranquillitas Pl. 2,20-25). For the first time we have what is surely a deliberate ffort to produce agallery of virtutes. The impression s of a ruler possessed of endless virtues.65

Why this sudden upsurge of interest n virtues? There can be little hesitation

in identifying he context. At the turn of the century Pliny had published hisPanegyric, lattering he ruler not as a god but as a man.66 Twenty perfectlyhuman virtutes, moral qualities, are mentioned within three chapters alone (2-4), and at least fifteen more in the rest of the speech.67One may well supposethat earlier gratiarum aiones praised earlier mperors or virtues: but it is nomere accident of transmission hat ensured he survival f Pliny's.68 he title of'optimus princeps', redolent of the philosophical deal of the rule of the best

62 I have argued the case in detail in NC 1981, 20ff.63 J. Beaujeu, La religion romaine a I'apogee de l'empire 424 noted the frequency of Hadrian's

innovations in this area.64 The point is well made by P. G. Hamberg, Studies in Roman ImperIal Art (1945), 32ff.65 For this 'gallery' of virtues see Strack ii, 123ff.; Mattingly BMC iii, cxl. Mr. E. L. Bowie

points out to me that the 'series' is a feature of Hadrianic coinage, comparing the Provinces series.This helps to explain why there was no earlier 'virtue series'; but the point remains that Hadrianshowed an unusual interest in both Provinces and virtues.

66 Pan. 2,3 'nusquam ut deo, nusquam ut numini blandiamur'; etc.67 These are: pietas, abstinentia, mansuetudo (2,6); humanitas, temperantia, facilitas (2,7);

pudor (2,8); modestia, moderatio (3,2); frugalitas, clementia, liberalitas, benignitas, continentia,labor, fortitudo (3,4); severitas, hilaritas, gravitas, simplicitas (4,6). Add later vigilantia (10,3);indulgentia (21,4); bonitas (30,5); iustitia (33,2); veritas (54,5); patientia (59,3); sanctitas (63,8);fides (67,1); reverentia (69,4); comitas (71,6); aequitas (77,3); diligentia (92,2). (Only earliestoccurrences given.)

'8 For earlier gratzarwm actiones see the edition of M. Durry (1938), 3f. Pliny's Panegyricsurvives because it was used as a model by later panegyrists: it was evidently an outstandingperformance and treated as such at the time (Plin. ep. iii, 18).






emperor's possession of virtues becomes a cliche.72 If the aim of the coinage

was to spread belief in the emperor's virtues, it was not until the 2nd centuryAD that it achieved its purpose.

3. Virtues and Powers

But the distinction between personifications that represent moral qualitiesand the rest, though it may serve to make a point, is not an illuminating one forthe understanding of the coins themselves. 'Virtues' is a confusing termbecause it coincides with the ancient term virtutes which has rather differentconnotations. But even if the label is unfortunate, the category of personifica-tions which it describes is indeed one that hangs together. The question mustbe approached again of what these personifications are, how they are distinctfrom the qualities met in Pliny or Suetonius, and where there is overlap.

One may start by asking how a Roman would have explained the functionof numismatic personification. Arnobius, at the beginning of his fourth bookAgainst the Gentiles, attacks the pagan Romans for their needless multiplica-tion of divinities.73 Pietas, Concordia, Salus, Honos, Virtus, Felicitas are madeobjects of cult, when they are no more than blessings we pray for; do they

genuinely believe that Victoria, Pax and Aequitas live up in the heavens, or isthis a faqon de parler? Arnobius then puts his finger on the truth: it is acharacteristic trait of Roman religion to turn abstractions into numinouspowers. He opens his copy of Varro's Antiquities, evidently under the letter P,and is able to reduce this whole tendency to absurdity: Panda is the goddesswho opened a way up to the Capitoline for Titus Tatius, Pellonia the one whodrives off the enemy. Worse, Pertunda is the goddess of sexual penetration,and Perfica of sexual performance. Prayers may have become more sophisti-cated since those old days, but the mentality is the same.

Now it is clear that as a Christian polemicist Arnobius is pushing the paganinto an extreme position. If we ask whether an intelligent pagan would havebelieved these personifications were literally gods, on a footing even with theOlympians, the answer is certainly that they did not. Arnobius' polemic isbuilt on the basis of earlier pagan polemic, as for example, in Cicero, Pliny theElder and Lucian, against the 'hypostatisation' of abstractions."4 However,even their polemic implies that others, more naive, believed. Another approach

72 The earliest attestation of such superlative virtue appears to be Fronto de Fer. A/s. ii, 6 p. 215v. d. H. of Pius 'omnes omnium principium virtutes supergressus'. ILS 374 has the samephraseology of Marcus, but glorias in place of virtutes. Cf. ILS 400 (Commodus); 597 (Probus);733 (Constantius).

7 Adv. Nationes iv, 1-12. This is a standard topic of Christian polemic; see Augustine Civ. Deiiv, 21 and passages cited by Pease on Cicero ND ii, 61.

" Cic. ND ii, 61; Plin. NH ii, 14; Lucian Conc. Deorum 13.




is to ask whether hese personifications were actually he recipients f attestedcult.75 Some enjoyed state cult in Rome (Concordia, Spes, Pietas, Pudicitiaetc.) or received sacrifice from the Arvals (Securitas etc.), or at least receivedprivate dedications (Aequitas). But many others, including so vital a virtue asLiberalitas, enjoyed no known cult.76 Scholars therefore distinguish these asdeities only in an 'allegorical' sense: they are personified (or hypostatised) onlyfor the purposes of numismatic picture language.77 But here again it is pointlessto split the coins into two classes. When Liberalitas and Clementia both wererepresented as female figures with divine attributes, it meant nothing to thecoin-user to know that one had a temple at Rome, the other not.

Numismatists therefore tend to abandon the fact that these are deities, andtreat them all as abstractions. Yet there is a danger in totally secularising them.The obverse of the coin was occupied by the head of the reigning emperor, aposition reserved until the time of Caesar for real divinities. Of course nobodysupposed that this implied that the emperor was a god ;78 even so it was one ofthe numerous signs that he occupied a place of ambiguity between humanityand divinity. The reverse bore an evident connection with the obverse; and inthe case of personifications it was frequently (but not always) emphasised bythe attachment of the label Augusti/Augusta/Augustorum. The personification

was therefore in some sense in the power of, or an aspect of the emperor on theopposite side. Seen therefore from the exaggerated viewpoint of an Arnobius,the message of the coin was roughly this: if you want peace, you will have topray to the deity of Peace; but Peace is in the power of Augustus, so you mustfirst pray to him.

This is of course to take the coins at face value. We need not suppose theRoman actually 'believed' so much. But the fact that the 'metaphor' the coinsemployed was a religious one is in itself significant. It identifies the view of theemperor as a 'charismatic' one. The subject is encouraged to attribute to his

sovereign qualities that are (at face value) supernatural. The emperor owes hisposition not only to the possession of a legal titulature (spelled out on theobverse), but to his possession of powers and qualities. In his hands lie Peace,Concord, Felicity, Security, Safety, Trust, Good Fortune and the like; they areguaranteed by his Victory, secured through his unique possession of Virtueand the favour of the gods given to Piety. Other material benefits are under hiscontrol, the Corn-supply, even Money itself; they are made possible by his

For the details Wissowa, Religion und Kaltus 327ff.; Beaujeu, La religion romaine 424ff.;Latte, Rom. Religion 321ff.; J. Bayet, Histoire de 1a religion romaine (1969), 109ff.

76 Kloft (1970), 181.77 So J. M. Toynbee, 'Picture-language in Roman Art and Coinage', in Mattingly Essays 205ff.;

esp. 216ff.78 Thus the distinction drawn by Christ a propos a Roman coin: 'Render unto Caesar the

things that are Caesar's, and unto God . . .' Matt. 22,21 etc.




Providence, his Liberality, his Equity.79 Power, in fact, is at the focus of

attention: the power to conquer, to save, to bring harmony and stability, andto distribute benefits. This power, beyond the reach of the ordinary human,tends to the divine. As Cicero puts it (ND ii, 61) humans turn abstractions intodeities because of the power inherent in them, 'quia vis erat tanta ut sine deoregi non posset'.

The basic message of the coins, therefore, is not that the emperor has theright human qualifications to suit him for the job, but that he has the necessaryalmost mystical powers and gifts (or that he enjoys the divine favour required)to grant his people what they needed. This was essentially at odds with the

'rational' approach that characterised the educated: only the morally best manwas fitted to rule, and only virtue in the moral sense, not power or wealth offortune elevated men above the level of the human. One has only to enumeratesome of the virtues attributed to emperors in the Panegyric or the Caesars tosee that they belong to a different world of thought: abstinentia, moderatio,continentia, humanitas, civilitas, comitas, facilitas, simplicitas, veritas,frugalitas80 are enough to give the flavour of what is missing. These are aboveall social virtues, qualities of self-restraint. The focus is not on the possessionof power, but on the control of it in deference to other members of society.

Again one must not make the distinction too rigid. Inevitably there is adegree of overlap. 'Rationalising' writers know the value of victory, fortune,security, concord and the rest; and the coins, particularly under Hadrian,make some attempt to advertise more human virtues (note especially Patien-tia). In particular the two great 'ideals' of Clementia and Liberalitas are (moreor less) prominent in both coins and literature. As is also true of Libertas, theyhad become political 'slogans' of far too much emotive resonance to beignored. These are themes of central importance for the understanding of theempire, and it makes sense to gather together the evidence of these 'ideals'

from different sources (as do Wickert and Kloft). But that is not to say thatcoins and literary sources make the same use of these slogans. If emperors hadfollowed the moral rules laid down by Cicero in the de Officiis for the exerciseof Liberality, this source of political power would have been gravelyweakened.8"

A second caveat is that no simple contrast can be drawn between 'elite' and'masses'. Doubtless, as Charlesworth suggested, the coinage reached a widersocial range than Seneca's de Clementia or Pliny's Panegyric. It is also fairly

79 I do not mean to imply any exclusive connection between these benefits and these qualities.80 cf. above n. 67 for Pliny. For Suetonius ee conveniently W. Steidle,Sueton und die antike

Biographie 1951), 112. The virtues he persistently requires of an emperor are abstinentia,moderatio, iberalitas, clementia and civilitas. For a recent analysis, E. Cizek, Structures tideologiedans es Vies des Douze Cesars de Suetone 1977).

81 On Cic. de off, 1, 42ff. & ii, 52ff, see Kloft (1970), 39f.




clear hat those who stood to benefit most from the virtues of restraint tressedby literary sources were the educated upper classes. But there could be nojustification or arguing that coins were aimed exclusively at the masses.82More important, the elite used precisely the same semi-religious anguagewhen occasion demanded. As Arval Brethren, eading senators offered actualsacrifice on behalf of the emperor to such abstractions as Aeternitas,Concordia, Felicitas, Fortuna, Providentia, Salus, Securitas, pes and Victoria;as also to the Genius of the princeps.83One cannot even draw a clean linebetween iterary and epigraphic ources. There s a contrast between he imageof Trajan n the Panegyric nd in the Letters of the same author. When Plinywrites to the emperor he invokes his Aeternitas, nd celebrates is Providentia:in the Panegyric we hear neither of Eternity nor of Providence.84 n factdifferent contexts demanded different language. The real contrast is notbetween social strata elite v. masses), nor between media (coins v. literature)but between two different aspects of the emperor that may be labelled the'rational' and the 'charismatic'. hey are two different ways of looking at theemperor that may overlap even within the same or similar contexts. Particu-larly in the late imperial panegyrics he two approaches re mingled, thoughsome, especially Mamertinus' anegyric of Julian delivered before the senateof Constantinople) mitate Pliny's in their rationality, while others stress themystical and charismatic ide.85

Conclusions

It is now time to summarise he results of this investigation, and ask again'What are Imperial Virtues '. The argument moved from Charlesworth'shypothesis: virtues provided a charismatic ustification of the emperor'spower, representing im as in possession of qualities regarded y his subjectsas a necessary qualification or his position. So much may provisionally beaccepted. Charlesworth hen went on to attribute a key role to a generally

82 Nor can any real contrast be discovered between the message of the precious metals as forthe elite, as against aes for the masses, pace T. F. Carney, The Turtle (N. Am. Journ. Num.) 6, 1967,291ff.

8 See Henzen, Acta Fratrum Arvalium (1874) Index s. v.84 Pin. ep. x, 41,1 for the invocation of Aeternitas; for providentia/providentissimus 54,1;

108,2; 61,1; 77,1. Providence in the Pan. belongs to the gods (10,4); 'provida severitate' 34,2 is adifferent use of the word. Providentia and aeternitas are shunned by Tacitus: see Syme, Tacitus754f.

85 See F. Burdeau, 'L'empereur d'apres les panegyriques latins', in F. Burdeau, N. Carbonnel,M. Humbert, Aspects de l'empire romain (Paris 1964), 1ff. I owe much to this excellent paper informulating this distinction.




agreed 'canon' of virtues. It is here that his case not only founders on theevidence but leads to a grave distortion. The idea of a canon provides a linkbetween different and possibly conflicting viewpoints: Greek philosophicalreflections on the duties of a man, and particularly f a king; the pressuresbrought to bear by the upper classes, notably the senate, on individualemperors to conform to a pattern of behaviour that was to their ownadvantage; nd the possibilities of official persuasion or 'propaganda' fferedby the imperially ontrolled media, especially coinage.

Reduction of the 'virtues' met in these sources to a homogeneousconglomeration has the effect of depriving hem of their value as historicalevidence. Each mention f a virtue nly adds to a vast and ultimately unrealfiction of the 'ideal ruler'. This is not only uninteresting but chimerical.Rather, I suggest, we should look at the way that individual ources adaptgeneral assumptions hat the ruler should be virtuous to their own purposes.Tentatively offer the following as the broad outlines.

Greek philosophy played an important part n stimulating he use of virtuelanguage; hough one must also remember he role played by the tradition ofhonorific decrees, first developed in the Athenian assembly n the late 5thcentury, and on the other side the Roman aristocratic edication o virtus. For

the educated elite of the empire, to which of course the emperor himself(usually) belonged, philosophy provided a framework within which autocracycould be justified at a rational evel: the best and most virtuous man wasentitled o rule his inferiors. Philosophy did not, however, dictate he choice ofcriteria, he range of virtues demanded, ither on the coinage or in any othersource except the encomia prescribed by schools of Greek rhetoric.

From the writings of the Roman senatorial and equestrian elite, exemplifiedhere briefly by Pliny's Panegyric and Suetonius' Caesars, we should not expectto extract a universally valid ideal. Their use of virtue language should

illuminate the points at which they felt threatened: where the bad emperorcould damage their interests, and the virtuous one be prevailed upon to respectthem. I suggested above that the virtues of the Panegyric were those of self-restraint, of conformity to the interests of society as defined by the speaker.The justification for the emperor's possession of power becomes his willing-ness to abstain from using it to the detriment of those concerned. I shall argueelsewhere that the key points of concern are the protection of property, ofpersonal security (life and death), and of social standing.

The elite were little concerned with the justification of the system of

autocracy. It was accepted as a fact of political life that this was the onlycondition under which stability was possible. What mattered was the conductof the individual ruler, the use to which he put his inevitable power. But theimperial coinage has exactly the opposite emphasis. Once the language ofpersonifications comes into its own (after the civil wars of 68/69, reviving the




language of an earlier period of civil strife), then the themes used recur fromreign to reign. There s little concern with advertising he personal attributes fthe particular uler; attention s focussed on the benefits of autocracy tself: onthe possession of power, begotten of military victory, and of the stability andprosperity that derived from it. The Imperator possessed power analogous tothe divine; the personifications identify the various aspects of use of power,the patron saints, so to speak, of various benefits, under the ultimate control ofthe emperor. It is a development of the 'Golden Age' of the Antonines, thatwhat the ancients called virtutes came to occupy a regular, but minor niche inthis Pantheon. So much is surely a response to the vociferous virtue-talk of theelite, of which Pliny's Panegyric provides the primary evidence: yet the virtueshenceforth celebrated are not the type on which a Pliny laid emphasis, butthose that conformed to the general message of the benefits of autocraticpower.

Where and why there is overlap between the different points of view is aquestion that demands further investigation. But it cannot be taken forgranted, or explained by reference to a 'canon'.

Magdalene College, Cambridge Andrew Wallace-Hadrill

APPENDIX

Personifications on the official coinage

A similar tabulation will be found in F. Gnecchi, 'Le personificazione allegoriche sulle moneteimperiali', Riv. It. Num. 18, 1905, 349ff. The following one has been constructed independently,on the basis of the Indices of Mattingly's BMC Emp. volumes, and is in several respects different.Reigns have been divided into three periods (see text). Further, personifications have been dividedinto different classes. These are, from a numismatic point of view, arbitrary, and serve only tomake easily visible the points made in the text. The separate classes are: i) the so-called 'canon ofvirtues' of the Golden Shield; ii) personifications that from a philosophical point of view mayreasonably be termed 'virtues'; iii) Aequitas and Moneta correlated in order to show theirinterdependence; iv) other personifications.

Such tabulations inevitably conceal important facts. I have not distinguished different types, thefrequency of individual types within reigns, nor types which appear against an obverse bearing the

head of a member of the imperial family other than the reigning emperor. Such points are not hererelevant. Nor is any indication given of whether types are identified as AUG(usta/usti/ustorum)or P(opuli) R(omani) since practice varies frequently from reign to reign, and even within reigns(see below on Tranquillitas). I have however attempted to indicate one distinction. A personifica-tion strictly is the figure of a deity (usually female, but Honos and Bonus Eventus are male;Genius has been excluded); normally identifying legends are attached to these. Where they are not,




there is danger of mistaken dentity; these are recorded as T(ype). (E. g. It is impossible odistinguish he Type of Aequitas rom that of Moneta.) Sometimes owever, he legend names anabstraction, ut it does not accompany personification, ut rather a scene, an altar, a shield orsuch like. It is vital not to confuse hese(e. g. Tiberius' Clementia nd Moderatio ppearing n ashield are presumably he record of an honour) and they are distinguished s S(cene).

Since the object is not to show the choice of types by individual mperors but to establishoverall patterns, he data for the period between Pertinax nd Diocletian has been reduced o anapproximate ndication of frequency within the bands R = 5-15, F = 16-25, FF = 26-35,FFF = almost without exception; figures ess than five are given precisely the unit being theemperor or claimant under whom the type appears). No indication s given of personificationsintroduced ater than Commodus; n fact these are rare notably Abundantia rom Elagabalus,Perpetuitas rom Severus Alexander, Uberitas from Trajan Decius). The HUMANITAS AUGattributed o Probus RIC v, 2, 36, cf. p. 7) is highly dubious. Brief comments n selected pointsare given below; for full details and discussion ee Mattingly n BMC -iv; Strack -iii; and whereappropriate Sutherland 1951); Grant (1950). On personifications ee briefly Grant (1953)133-175.W. Koehler, Personification bstrakter egriffe u/frm. Munzen, Diss. Konigsberg 910only reaches lementia roceeding lphabetically; ut note the brief posthumous ontributions fthis author o the Enciclopedia ell'Arte Antica (1958-66). For the Republic, M. H. Crawford,Roman Republican Coinage 1974).

I) virtus: Not a true type of Augustus since only under the independent moneyer AquilliusFlorus, (BMC , 10) reviving a type of an ancestor (RRC no. 401). It is symptomatic of the lack ofdesire to advertise he virtues of the Golden Shield, or indeed Virtues n general, hat Virtus,though a republican heme, does not appear before Nero: (Sutherland 951, 159).Even hen theiconography makes t clear hat the connotations re strictly martial.

clementia: First as the temple o Clementia Caesaris RRC no. 480, 21). Then under Tiberius nthe much-debated win series with the legends CLEMENTIAE and MODERATIONI(S)respectively ound imagines lipeatae. Date and occasion of this series s still not settled, but itclearly ommemorates ome honorific dedication, nd must not be equated with 'personification';see Sutherland, RS28, 1938, 129ff.;Grant, 1950,47ff.; Sutherland 951,193ff.;H. Gesche,Jahrb.Num. Geld 21, 1971,37ff.; B. Levick n Essays n honour of C. E. Stevens 1975),123ff.The type

enjoys a brief lowering under he Antonines, but thereafter s exceedingly are Clodius Albinus),except under he rather different guise of CLEMENTIA TEMPORUM Gallienus n).

iustitia: For full details ee now Lichocka 1974).Of Vespasian's ype (BMC i, 75) there s onlyone attestation, nd that specimendisappeared nder he French Revolution Lichocka 2, n. 25).It must therefore come under extreme suspicion. The iconography, eated goddess with ear ofcorn, has not been hitherto correctly explained: t is the astrological ign of the Virgin VirgoSpicif ra). This increases ts oddity. For Trajan wo types come under consideration; ) BMC ii,lxv goddess seated with branch nd slanting ceptre, aken by Mattingly following Strack , 52)asboth Pax and lustitia, by Lichocka 4f. as Pax alone; ii) BMC ii, lxvigoddess enthroned oldingsceptre with cornucopiae as arms; Strack , 65f. identifies as 'Secura Annona'; Mattingly asJustitia-Astraea, he 'spirit of the Golden Age'; Lichocka 44ff. as Justitia. . e. We can identify

neither igure.pietas: The popularity of this traditional Roman theme needs no comment. For the strong

charismatic ssociations f this virtue ee Charlesworth, Pietasand Victoria: he Emperor nd heCitizen', JRS 33, 1943, Iff. On imperial oinage of the 1st century AD the reference s almostexclusivelydynastic: Strack , 75; ii, 51f.; 169f.For further bibliography Weinstock 1971),248n. 2.



PLATE 1

0 0

-e

.fl

7 1

_iFl A Y W >~~~~~~~~~~oxd * Y~~~~~~~~~~~~~~~~

s* r & F1~~~~~~~~~~

$ > -a a D~~~~~~~~~~ri

+n'''6vp}m=E;~~~~~~~~

iw L{/Y i \+s#t #i Ss0-d _K ^ w es G~~~~~~~~

_ >g C:~~~~~~~

a .t~~~~~~~~

.v ^ '/ v~~~~~s ^~~~~~~~~~~'

, - =~~~

'. .': O~~~., ._ , ' r~~~~



PLATE 2

1 2 3

4 5 6 7

8 9 10 11

12 13 14 1S

16 17 18 19

20 21 22

23 24 25

'Virtues' n Coins




II) ('> indicates that classification as a virtue is questionable.)constantia: Exclusive to Claudius: M. Grant, 'Constantia Augusti', NC 1950, 23ff. exaggerates

vastly in characterising the virtue as an epitome 'of the whole current trend of thought concerningmonarchy'.

*disciplina: Not properly a virtue (its placing in the chart is dubious). Celebrates Hadrian'sencouragement of the military cult of Disciplina (Strack ii, 151f.).

liberalitas: he clearest case of the elevation of concrete circumstance (the congiarium) to anabstraction. Trajan tentatively introduces the goddess (unnamed) on aurei (Strack i, 141f.),Hadrian both introduces an identifying legend for this personification, and uses the same legend toreplace CONG(iarium) P. R. The widespread epigraphic celebration of the abstract Liberalitas iswell illustrated by G. Barbieri Diz. Ep. iv, 838ff., s. v.

'*magnificentia: Not a personification (MAGNIFICENTIAE AUG within laurel wreath);presumably, like Moderatio, a reflection of an honorific dedication. AD 192 only.

moderatio: Not a personification: Tiberius only. See above clementia.providentia: The only type between Tiberius and Vitellius is an altar, PROVIDENT. S. C.,

evidently the Ara Providentiae Augustae dedicated under Tiberius (Eisenhut RE Supp. xiv (1974),562f., s. v.). This is no personification, and hardly suggests the current emperor's possession of a

KEY TO PLATE 2

1. Tiberius Rev. IUSTITIA BMC i, 131, 79; Fitzwilliam

2. Tiberius Rev. PIETAS BMC i, 133, 98; Fitzwilliam3. Tiberius Rev. SALUS AUGUSTA BMC i, 131, 81; Fitzwilliam4. Tiberius Rev. CLEMENTIAE BMC i, 132, 85; B.M.5. Tiberius Rev. MODERATIO BMC , 132, 90; B.M.6. Gaius Rev. Three sisters as goddesses BMC , 152,36; Fitzwilliam7. Claudius Rev. CONSTANTIAE AUGUSTI BMC , 184, 140; Fitzwilliam8. Libo Obv. BON(us) EVENT(us) RRC416,1; Fitzwilliam9. Paullus Lepidus Obv. CONCORDIA RRC415,1; Fitzwilliam

10. Brutus Obv. LIBERTAS RRC 433,1; Fitzwilliamll. Mn. Aquilius Obv. VIRTUS RRC 401,1; Fitzwilliam12. Civil War, AD 68 Obv. BON EVENT BMC i, 289, 16; Fitzwilliam Archive

13. Civil War Rev. CONCORDIA PRAETORIANORUM BMC i, 305,61; B.M.14. Civil War Obv. LIBERTAS RESTITUTA BMC i, 292,12; B.M.15. Civil War Obv. VIRT(us) BMC i, 295,18; B.M.16. Titus Rev. BONUS EVENTUS AUGUSTI BMC ii, 241, 106; B.M.17. Galba Rev. CONCORDIA (PROVINCIARUM) BMC i, 348, 225; B.M.18. Galba Rev. LIBERTAS RESTITUTA BMC i, 339, 177; B.M.19. Galba Rev. VIRTUS BMCi, 342, 195; B.M.20. Hadrian Rev. CLEMENTIA AUG BMC iii, 304, 513; B.M.21. Hadrian Rev. INDULGENTIA AUG BMC iii, 305, 521; B.M.22. Hadrian Rev. IUSTITIA AUG BMC iii, 305, 522; B.M.23. Hadrian Rev. LIBERALITAS AUG BMC iii, 305, 523; B.M.

24. Hadrian Rev. PATIENTIA AUGUSTI BMC iii, 306, 525; B.M.25. Hadrian Rev. TRANQUILLITAS AUG BMC iii, 306, 526; B.M.

My thanks are due to the staff of the British Museum and of the Fitzwilliam Museum, Cambridgefor their ready assistance in producing t'hese photographs.

21




virtue. From Vespasian n is found an interesting eries of scenes referring o the accession; hegoddess only appears personified under Trajan, nd is labelled by Hadrian Strack , 45ff.; 228).For Providentia s an imperial logan Charlesworth, Providentia ndAeternitas', HThR 9, 1936,107ff.

pudictia: A coin of Plotina under Trajan epictsan ARA PUDIC(itiae), BMC ii, 107. Hadrianintroduces he personification, MC ii, cxxxi.Mattingly loc. cit.) unconvincingly dentifiesP. asthe personal holiness of the head of state religion; t is almost always a theme for the coins offemales of the imperial house who are taken to embody this traditional Roman women's deity(Strack i, 117f.). To see in the type a counterblast o rumours of Hadrian's personal eroticexcesses,as does Carney op. cit. (n. 82), 291ff., strains redulity.

*tranqlitas: Closely akin to Hilaritas, ntroduced n the same reign. In neither ase does itmake sense to distinguish he 'virtue' rom the 'res expetenda'. TRANQUILLITAS AUG mightbe taken as the emperor's philosophic peace of mind' (Mattingly ii, cxl, n. 7); but it is also thedisposition n the emperor hat produces he 'tranquillitas aeculi ui' (Plin. ep. x, 3A, 2, cf. Stracki, 124f.). HILARITAS P. R. is more ostensibly a 'res expetenda' therefore not here classed as avirtue), but it is undoubtedly onceived of as produced by the Hilaritas f the emperor cf. Plin.Pan. 4, 6). Mattingly ii, cxxxiiiaptly citesFronto ad M. Caes. , 9, 7p. 20, 3f. v. d. H. 'certe hilarises .. . mea securitas, ilaritas, loria'. Tranquillitas nly returns under Philip I (RIC iv, 3, 63) andTacitus RICv, i, 342).Hilaritas scommoner, lsoas HILARITAS AUG and HILAR TEMPOR.

III) aequitas/moneta: or the close interrelationship f these types see Mattingly BMC iv, 1.Against he reference f numismatic Aequitas o the administration f justice by Strack , 154ff.)For detailed arguments ee NC 1981, 20ff.

IV) concordia: A central hemeof imperial deology, and after Victoria he commonest ype. See J.Beranger, Remarques ur la Concordia dans a propagande . .', Festschrft F. Altheim (1969),477ff.

felcitas: the civil war type is ill-attested BMC , 290 n.) but I accept t as fitting he pattern frepublican ypes (RRC no. 473,3) later aken on by Galba.

fortuna: Under Augustus only a commemoration f the altar Fortunae Reduciby the moneyerRustius BMC i, 1).

hilaritas: . s. tranquillitas.honos: Under Augustus only the moneyer Durmius BMC i, 10. Later always n connection

with Virtus, as in republican ult. M. Bieber, Honos et Virtus', AJA 9, 1945,25ff.; Eisenhut RE

Suppl. xiv,905f.

Galba,ollowed

byVitellius and

Vespasian, airsHONOS ET VIRTUS, as the

republican RRC no. 403 (70 BC); Pius has two separate ypes of Honos and Virtus or Marcusperhaps as princeps uventutis BMC v, lviii).

victoria: The commonest of themes, n innumerable arieties. Rightly so, since Victory s thelynch-pin of imperial doctrine: J. Gage, 'La Theologie de la Victoire mperiale', Rev. Hist. 171,1933, 1ff.Already egular nder he republic with the adoption f the Victoriatus' s normal ypefor gold quinarii




t x | x Zxx x x x x n 0 0 I>xVRU

~~~~~~ ~ 04-4

Z c

z~~~~

~xxI x X X XX X XIH0- I XVIRTUJSx-____ _ ___ __ _ _

__________

Pd XIX xX I _'CLEMENTIAP x X X X--I x IUSTITIA

XI X X X X X -X X PIETAS

O _I_ _ _ _ _ _ _ _ __ l x I _ CONSTANTIA

0 1 C^ I *DISCIPLINAO I I x xXX INDULGENTIA

Ix x I xx ___ LIBERALITAS

ocun _ I *MAGNIFICENTIAO I - _ MODERATIO

X I x I MUNIFICENTIA0 x I PATIENTIA

Xx ix I Iw _ _ _ _ I PROVIDENTIAw x xI X _ PUDICITIA

I X I *TRAN UILLITAS

xx jX t t tQiXXXX XXIX -X X I _ AEQUITASIXH,-IX X XqxxI I XMONETA

OIT1 XI XX X XX I I AETERNITASI'Xl HI x1x H X XXx XX I X ANNONAt--HH --I x _x _X _ xx I I x BONUS EVENTUJS

I x -XX

I3 x x x xx x Ix H

ix CONCORDIA15 x x I x INI I I FECUNDITAS

W l<xx x x XX _ x FELICITASTI x xIx [x x x x x x x x I x FIDES[i x x_ x x x x x __x x _ _vixFR*N' x Ix xxxxx Xi I X FORTUNAXt x 1Xx I HILARITAS

?9 l x F -- I X HONOS17 X X I X I LAETITIA1NiX I XX X X X X X I X IX LIBERTAS

POXI--I--_ __I_______-- NOBILITAS1X XI x X X XX X XX X X I XU X I X PAXXVX XXX X X X X XIX SALUS

L X X I X i S C U IXXI XXH XXXXX H I I SECURITAS

12 X I X X H_ X I X _ IISPES

L; 1 __ e - rI _ IITUTELALX xXXx x x x x x x x x x x I xXHH HIIj qVICTORIA

21*

Wallace-Hadrill, A. - The Emperor and His Virtues. 1981

Documents