PHOENIX Caput and Colonate:

JOURNAL OF THE CLASSICAL ASSOCIATION

OF CANADA
TOWARDS A HISTORY OF LATE ROMAN TAXATION

RBVUE DE LA SOCIETE CANADIENNE

DES ETUDES CLASSIQUES

SUPPLEMBNTARY VOLUME XII

TOME SUPPLBMENTAIRE XII

WALTER GOFFART

.^
0^»
'^
UNIVERSITY OF TORONTO PRESS

£*s"--\"^~
) University of Toronto Press 1974 TO VIVIAN AND ANDREA,
Toronto and Buifalo
JOYFUL VOICES OF A DECADE
Printed in Canada
ISBN 0-8020-5289-4
LC 73-85096

EX LI^IS
UNIVERSiTATIS
NOVIOMAGENSIS

. 3 i'1.. 6^
 Contents

ABBREVIATIONS viit

PREFACE iX

INTRODUCTION 3

i Provmcial tribute under the early Empire 6

ii The setting oftaxation in the early
fourth century 22
m lugumandcapitatio 31
iv The three meanings ofcaput 41
v From caputto colonate 66
vi Conclusion 91
How was provincial land taxed in the early
Empire7 99
The exclusionofthe imperialgovernmentfrom
local economies 102

The effect ofiuga. and capita upon the integrity

oflocaleconomies 104
APPENDIX I: THE CENSUS INSCRIPTIONS 113

APPENDIX H: SUPPLEMENTARY NOTES 122

POSTSCRIPT 139

INDEX OF SOURCES 147

SHORT TCTU INDEX OF MODERN AUTHORS 150

GENERAL INDEX AND GLOSSARY 155

 ABBREVIATIONS Pref.ace

CJ Codexjustinianus,ed.P.Krueger,in CorpusIwisCivilis,3 vols. (reprint, This book was composed in several stages. The first draft ofwhat is now
Berlin, 1954-67), n chapters n to iv was written in the spring of 1969; chapters i and vi were
CTh Codex Theodosianus,ed. Th. Mommsen (reprint, Berlin, 1962) added in the autumn of 1972. I am still actively engaged in studying late
Dig. Digesta,ed. Th. Mommsen, in Corpus luris Civilis, I Roman taxes, andtheir impact onthe early Middle Ages, andoffer the present
FIRA2 FontesiurisRomaniantejustiniani,ed.altera,3 vols. (Florence,1940-3) book as a report ofresearch in progress.
h.t. hoc titulo, in the title just cited" Three institutions have supported my research: the Institute for Ad-
Nov. Novella {lex), particularly ofjustinian, ed. R. Schoell and W. Kroll, vanced Study, Princeton, gave me its superb hospitality in 1967-8; the
in Corpus luris Civilis, m Canada Council a Senior Fellowship in that same year; and the Humanities
NAnth, NMaj, NTheod, NValent Novellae of the emperors Anthemius, Research Board, University of Toronto, a grant for clerical assistance. It is
Majorian, Theodosius n, and Valentinianm, in Leges novellae ad Theo- a pleasureto thankthemfor their generosity.I amalso gratefulto Professor
dosianum pertinentes, edd. Th. Mommsen and P.M. Meyer (reprint, T. D. Barnes, University College, University of Toronto, for his assistance
Berlin, 1962) inreadingGreektexts,aswellasto Mrs. LindaWilding,formersecretaryof
P. papyrus; see Index ofSources the Centre for Medieval Studies, University of Toronto, who cheerfully
pr. praefatio,the openinglines ofthe law or legalextractcited andflawlesslytyped the manuscript. I hope the friendsand colleagueswho
TLL Thesawus Ungiiae Latinae (Leipzig, 1940- ) have kindly read portions ofthe manuscript at various stages and offered
advice will accept my collective thanks.
In footnotes, an asterisk at the end of a sentence indicates that further dis- Publication of this book has been made possible by a grant from the
cussion is to befound in Appendix 11(beginning on page 122), andanasterisk Humanities Research Council of Canada, using funds provided by the
immediately following a crossreference to another footnote indicates that the CanadaCouncil, and a grant to the University ofToronto Press from the
relevant portion of the note may be found in Appendix 11. Andrew W. Mellon Foundation.

The Postscript is indexedin the GeneralIndexand Glossary but not in the W. G.

Index of Sources or the Short Title Index ofModern Authors. Dumbarton Oaks, Washington, D.C.
October 1973
CAPUT AND COLONATE
 Introduction

In a footnote to his account of late Roman taxation, Edward Gibbon re-

marked: "The first twenty-eight titles of the eleventh book of the Theo-
dosianCodearefilled with the circumstantialregulations on the important
subject of tributes; but they suppose a clearer knowledge of fundamental
principles than it is at present in our power to attain. "' New evidence has
sincebeenuncovered;newmethodshavebeenappliedto theanalysisofthe
sources;a shelffulofspecialstudieshasaccumulated.2 The subjectneverthe-
lesscontinues to beaselusive asGibbonadmitted it to be. Ourknowledge of
'fundamental principles has made little progress, and our image of late
Roman finance is correspondingly vague, especially since repetition has
tended to turn assumptions into facts.
Several authors have already traced the historiography of late Roman
taxation; the story need not again be told. 3 Ifthe knots can be unravelled, it

I History ofthe DeclineandFallofthe RomanEmpire,ed. J.B. Bury (London, 1896-1900),

n, 191, n. 180.
2 New evidence: many papyri, the Syro-Roman Lawbook, the Brigetio Table, the nine
census inscriptions from Asia Minor and the Aegean. The Labour Theory of Value,
injected by Rodbertus into the discussion (1864), was long influential in the interpretation
oftheevidence(below,ch.rv, n. 45).A. Deleage,in LaCapitationduBas-Empire(Macon,
1945)1 sought clarity by surveying the regions of the Empire one by one. He realized
(p.22)thatoneapproachwasasyetuntried,namelytheanalysisoftheevidenceinchrono-
logical order. I endorse the view of P. R. L. Brown, stated in Economic History Review,
2ndser.,20 (1967), 329,"thatthestudyofLateRomansocietybeganinthesixteenthand
seventeenth centurieswith the study ofthe Romanlaw codesofTheodosiusn andJus-
tinian, andthat it islikely to endwith them. " On the papyri, which arenot fully taken into
account in the present study, see below, ch. in, n. 3. The census inscriptions are surveyed
in Appendbc I. In references to the laws, dates followed by the letter s are given in
accordancewiththerevisionsof0. Seeck,RegestenderKaiserundPSpste(Stuttgart, 1919).
3 F. Lot, L'lmpSt fancier et la capitation personelle sous Ie Bas-Empire et a t'Spoquefranqi ie,
 4 CapU t AND COLONATB INTRODUCTION 5

must be by renewed analysis ofthe sources rather than by a debate with the The fundamental shift in the forms oftax assessment that this investigation
commentaries that have criss-crossed the same terrain from decade to decade reveals will then be shown to have beenparalleled by a changein the legal
since Godefroy and Savigny. 4 In view of this, the present study sets out status ofpersons, thepassagefrom contractual tenancytotheboundcolonate.
primarily to establish definitions for the main technical terms of late Roman Finally, the study suggests some directions for rewriting the history oflate
taxation, definitions that will be consistent with all the texts in which they Roman public finance.
appear. The only methodological novelty introduced into the discussionis
thatofkeepingtrackoftime. Existingworksonlateimperialinstitutionsare
too often anachronistic in their analyses of legal sources, interpreting early
laws in the light oflater ones. 6 The elementary technique ofsetting out the
legal evidence in chronological order will be the one applied here.
This study begins with an examination of taxation in the earlier
Empire,focusedonthequestionwhethertributumwasa directtax. A second
step, also preliminary, will be to reconsider the setting of taxation at the
begmning of the fourth century, especially as it is disclosed by the laws
granting exemption to various categories ofpersons. Once the air is cleared
of old assumptions, the discussion proceeds to the heart of the problem,
namely to define and elucidate the technical terms iugum, capitatio, and caput.

Bibliotheque de 1'Ecoledes Hautes-etudes,Scienceshist. et philol. 253 (Paris, 1928),

pp. 1-13; Delfage, Capitation, pp. 5-22. Lot brought the historiographical survey up to
date in his Nouvelles recherches sur I'impot fancier et la capitation personnelk sous Ie Bas-
Empire, Bibl. de 1'Ec. des Hautes-etudes 305 (Paris, 1955), pp. 9-24.
4 On Godefroy and Savigny, Lot, ImpStfonder, pp. 3-5; Deleage, Capitation, pp. 6-7.
The current state of scholarship on the question is described by Raymond Monier et at.,
Histoire des institutions et desfaits sociaux des origmes a I'aube du Moyen age (Paris, n.d.
[l955]). Pp. 276-7: "Lesreformes fiscalesdu Bas-Empire ...presententles plus grandes
difficultes pour les historiens modemes qui ne paraissent pas encore etre arrives a un
parfait accord."
5 ThemostrecentexamplesareA.H.M.Jones,"Capitatioandlugatio," JournalofRoman
Studies, 47 (l95?)> 88-94, whose results are incorporated in Jones's The Later Roman
Empire, 284-602: A Social, Economic,and Administrative Survey (Oxford, 1964), hereafter
abbreviated LRE;J. Karayannopoulos, Das Finamwesen desfruhbyzantlnischen Staates,
Siidosteuropaische Arbeiten, cd. p. Valjavec, 52 (Munich, 1958), and the same author's
DieTheorieA. Piganiolsiiberdieiugatio-capitatiounddieneuerenAuffassungenuber
die Entwicklung der sozialen und finanzwirtschaftlichen Institutionen in Byzanz,"
Byzantinisch-NeuegriechischeJahrbilcher,19 (1966), 324-49,whereno accountis takenof
the substanceofJones'swork. The latest contribution to this subject is Andre Cerati,
Carac&re annonaireet assiettetie I'impStfancier au Bas-Empire (Aix-en-Provence, 1968);
reviewedbyA. Chastagnolin Latomus,30(1971),495-501.Cerati's major contribution
bears upon the question how long in thefourth century the annonain kind constituted the
bulk oftaxation (see below, ch. v, n. 54). His discussionofthe tax baseis somewhatin-
decisive and more traditional than mine.
 PROVINCIALTRIBUTE 7

I pectedtQ_be complex. When we encounter simple words relating to it, we

are best advised to assume that they denote complex things, whose intri-
caciesshouldnot be forgotten evenifthey cannever beunravelled.
As long asschematictextsareunderstoodto embodydrasticSimplifi-
Provincial tribute under cations, there is no harm in utilizing them to gain an impression of early
imperialfinance.A usefulpoint ofdepartureisthestatementofTaatusthat
the early Empire Augustus's Breviarium imperii contained anaccount ofthe imperial revenues,
tribitta autvectigalia, alongsidethat ofthe imperial expenditures, necessitates
qc, largitiones. 4' Portoria and tributa were "the revenues that sustained the
Empire."5Ofthese,vectigaliaor portorianeednot detainus; the words refer
to a complex oflevies whosenature is comparatively well knownand not
A recent author has remarked that "the system devised [under Diocletian] subject to dispute. 6 The difficult term is tributa,^ which refersto the^revenues
to assess and collect the annona was simplicity itself. "1 Such a view owes ofthe imperial government fi-om the prpyinces only.
moreto modernsummaryaccountsofthe "system"thanto anappreciation Traditional doctrines on these provincial levies leave one with the
ofhow complex taxationalwaysis. Wherever one turns, governments are impression thatpatchy evidencehasbeenfitted intoaninterpretation inspired
found to havedeviseda largeand sometimes bewilderingvariety ofmeans by the tax structures of modern states. 7 According to Mommsen, the
by which to raise revenue. Ptolemaic Egypt is a notorious example; the transition from indirect to direct taxation of the subject population is one of
African kingdom of Dahomey in early modern times had a similarly re- the organic institutions of the Principate and one of its most profound
sourceful and inexorable fiscality; and even a state as well run as mid-nine- innovations ... "8 Mommsen's conclusion was based almost entirely upon
teenth-century Britain drew its revenues from no less than five distinct cate- the fact that provincial censuses were conducted by imperial officials:9 "If
gories oftaxes.2 The "singletax," imaginedand advocatedby modern re- Augustus had, like Caesar, been satisfied to set a fixed tribute upon each
formers,doesnotappeartohaveeverbeeninstitutedinhistory.Literarymen community, he would not have needed [to spend] forty years [in conducting
and administrators sometimes reduce fiscal complexity to simple, straight- censusesin Gaul]."10 Schwahn'saccount in Pauly-Wissowadoes not assert
forwardformulaethatareaccurateenough, but only if theyareunderstood thatthe tributumoftheearly Empirewas"directtaxation";but theimplica-
for what they are - not images of underlying simplicity but schematic
portrayals ofa situation so complex as often to be beyond our abilities to p4 Tacitus Annales 1. 11.4. 5 Jiirf. l3. 50.
reconstruct in detail. Even such an ostensibly obvious distinction as that 6 Fora summaryaccount,F.F. Abbott andA.C.Johnson,MunicipalAdministrationin the
between "direct" and "indirect" taxes has posed formidable problems of Roman Empire (Princeton, 1926), pp. 122-7. The numerous judgments pronounced by
interpretation to the u. s. Supreme Court. 3 In short, taxariqn must be ex- the authors, especially on portoria, are inspired by modern economic ideas of dubious
relevance. A comprehensive account is given by S.J. de Laet, Portorium: Etude sur
I'organisationdouanierechezlesRomains,RijksuniversiteitteGent,Werkenuitgegevendoor
i J.L. Teall, The Age of Consfantine: Changeand Continuity in Administration and de Faculteit van de Wijsbegeerte en Letteren, fasc. 10$ (Bruges, 1949), who conjectured
Economy, Dumbarton Oaks Papers, 21 (1967), 33. that portoria supplied asmuch ashalfofthe imperial revenues in the early Empire (p. 448).
2 For Egypt, GabrielArdant, Histoire de I'impot, I, De FAntiquite au xvim siecle (Paris, 7 This is particularly evident in derivative works; e.g., C.E. Cullen, "The Roman
1971), pp. 85-106. Dahomey: K. Polanyi, Primitive, Archaic, andModern Economics, ed. Revenue System, " Washington University Studies, Humanistic Series, 4th. ser., 8 (1920-1),
G. Dalton (Garden City, N.Y., 1968), pp. 215-25. Britain: Robert Blake, Disraeli 233, "With the development of the Empire, the centralization peculiar to that form of
(London, 1967), pp. 329-34; these categories, such as excise, were broken down into government tended to establish uniformity m the system of provincial taxes. Not
several taxes and schedules. surprisingly, Cullen was unable to discern any substantial change in taxation as between
3 Commentary to u.s. Constitution, art. I, sect. 2, clause 4, in Edw. S. Corwin, cd., early and later Empire (p. 237).
The Constitution of the U.S.A.: Analysis and Interpretation, Sznd Congress, 2nd sess., 8 Th. Mommsen, Romische Staatsrecht(reprint, Graz, 1952), n, part 2, p. 1095.
SenateDocument no. 170 (Washington, 1953), pp. 317-20, esp. 319. 9 Cf. ibid., H, 4i6-X7. 10 Ibid., a, 1094.
 8 CapU t AND COLONATE PROVINCIALTRIBUTE 9

tion that it was is plain: "The main contribution ofthe provinces was the or change. In Mommsen's view, the scheme discernible under Diocletian is
land tax, tributum agri"; moreover, "alongside the land tax, a head tax, "nothing but the old assessment in its original form extended over the
tributum capitis,... wasalsolevied in certainprovinces... "n Hetoo attributed Empire, " while Schwahn stressed the uninterrupted continuity of the two
great importance to the imperial role in assessment. The two taxes, on land types of tribute on land and on heads. 17 If it were mdeed correct that the
andon heads, wereraisedin accordance with a censusconducted by imperial Empire, from its beginnings, had directly taxed the provincials, then the
officials;12 this was not all: "The basis for [Augustus's] new ordering [of only significant novelty attributable to Diocletian in respect to taxes would
provincial taxation] was afforded by the general survey {Vermessung} ofthe be the extension, or restoration after many centuries, ofdirect taxation to
soil, which finally made possible the preparation ofan adequate cadaster for
each individual locality. "13 These conclusions had also been those ofMar- On closer examination, these doctrines appear to be more firmly
quardt: asserted than sustained. Marquardt admitted that the evidence on early
imperial taxation is sparse. 18 Although the literary sources explicitly refer
... the measure by which Augustus laid the basis for a reform of tax to the pair tributa and vectigalia, we have no text bearing out the modern
conditions thus consisted ofthe introduction ofa provincial census, contention that the land tax, tributum soli, and the head tax, tributum capitis,
It differed from the communal census already existing in a part ofthe were the most important taxes levied by the imperial government. 19 The
provinces in that it established the contributive capacity (Leistungs- government's role in taking censuses was crucial to Mommsen's conclusion
fahigkeit) ofthe whole province divided into great tax districts, and that Augustus introduced direct taxation; yet he recognized that there is
not ofindividual cities, and was carried out by imperial officials for insufficientproofthattheearlyimperialcensuswascentralizedandperiodic,
the purpose ofthe tribute levy. 14 or that it wasevencentrally organized.20Iftherehadbeena Reichscensus,he
said, we would have heard ofit. The lesson he drew, however, was simply
In short, imperial officials gathered personal and property declarations from that the imperial government, unlike that of the Republic, deliberately
the provincials, and on the basis ofthe assessment thus compiled the govern- concealed the vital statistics ofits rule; the data were available to the govern-
ment levied tributa upon the peoples it ruled. ment, but they werenot publicized.21Theargumentis obviously weak;on
A discussion oflate Roman taxation cannot disregard the earlier period the other pointsaswell, the discussionis so seriously qualifiedasto put the
because the way the earlier situation is portrayed has considerable relevance wholeinterpretationindoubt.Directtaxationleavesmanytraces;thevolume
to one s conception ofhow matters stood under Diocletianand after. Mar- oflate imperiallegislationon tHesubjectbearseloquent witnessto this fact.
quardt, for example, believed that the evidence on earlier taxation is supple- The sparsity of early imperial evidence is in itselfa reasonto reconsider£<
mented and clarified by the further development oftax legislation from Dio- whether the provinces really were directly taxed.
cletian onwards. 15 In other words, he was satisfied to infer earlier conditions The defects ofCurrent accounts are perhaps epitomizedby the state-
from later evidence. Schwahnexplicitly didsoin hisaccount ofheadtaxes. 16 ment that "the two forms oftribute" were the most important taxes levied
Another consequence was the approach taken to the question ofcontinuity
17 Mommsen, Staatsrecht, m, 229 n. i; Schwahn, "Tributum, " col. 54.
ll W. Schwahn,art. "Tributum," in Pauly-Wissowa,Real-EncyclopSiSieder classischen l8 Staatsverwaltung, ll, 217.
Altertumswissenschaft,2. Reihe,vol. 7A (1939), coll. 10, 11. 19 Trihifta and vectigalia: Tacitus Ann. 1. 11.4, 13. 50; Suetonms Vespasian. 16; S. H. A.
13 Ibid.,coll. 12-13.Alongthesamelines,A.H.M.Jones,TheGreekCityfrom Alexander Antonin. Pills 7. 8, rationes omnium provinciarum adprime scivit et vectigalium.
to Justinian (Oxford, 1941), pp. 138-9. Modern statements on tributum soli and tributum capitis: Schwahn, as above, n. ll; A. H. M.
13 Schwahn, "Tributum," col. 62. Jones,"Capitatio," p. 94; C.E. Stevens,"AgriculturalandRuralLifein the LaterRoman
14 J. Marquardt,Romische Staatsverwaltung,v. (Leipzig,1876),211. Empire, " in Cambridge Economic History ofEurope, 1, 2nd. cd. by M. M. Postan (Cambridge,
I5 Ibid., p. 217. 1966), ii3; F. Millar, The Roman Empire and Its Neighbours (London, 1967), p. 92;
l6 Schwahn, Tributum, " col. 68, where fourth-century laws are repeatedly cited to Ccrati, Carac. annon., p. x.
document supposedlyearly imperial conditions. 20 Staatsrecht, n, 1016, cf. n, 417. 21 ftirf., 417.
 10 CafU t AND COLONATE PROVINCIAL TRIBUTE II

by the imperial government, and By its corollary that the provincials paid indifferent to the replacement oftributum capitis by an endowment. 25Never-
'land tax" and "head tax" to the emperor. What is at issue here is not theless, the inscription forces us to qualify and refine our ideas of how the
whether the provincials paid direct taxes (among others) to some authority imperial government looked upon the "two forms oftribute. Instead of
or whether the imperial revenues included the proceeds ofsuch taxes Both defining them as imperial taxes collected from provincials, we are better
arebeyond question. But hasthetaxprocess beenaccurately portrayed when advised to interpret them as forms of direct assessment that the imperial
only this is said?22Isit correct to state, or evenimply, thatthetaxpayers were governmentauthorizedlocalgovernmentsto imposein orderto meetcom-
directly linked to the imperial government? The indispensable complement munal - not indiYidu^ to Rome.26 The two forms ofdirect
thatisoften omittedissomereferenceto therolepflpcal goYernmentasthe assessment weretoo lucrative, andtoo open to abuse, for the imperial govern-
intermediarybetweenthetaxpayersandthe emperor. 23 ment to allow the municipalities a completely free hand with regard to
When sufficient attention is paid to the municipalities in the tax them. 27 Even so, tributum soli and tnbutumcapi tis existed not as taxes that the
process, onecomesto questionwhethertheinterest ofthe imperialgQyern- imperial government collected but rather as forms of direct taxation by
ment resided in taxing - that is, in the precise devices by which revenue was which localities could raise the amounts they owed to the imperial accounts.
raised-or only m the total mmsth^ For example, a Local governments were not simply passive instruments within a general
benefactor in Tenos established an endowment to pay the tributum capitis of empire-wide tax scheme; they retained a substantial measure of freedom in
his community. 24 The replacement of a tax by an endowment shows, to thewaystheywentaboutraisingthetotalstheywereresponsiblefor.28The
begin with, that the (nfeM (Mm in qyestion grod^^^^^ that did not
evidence for local and provincial diversity in early imperial taxation is
varyfi-pm year toyear. The effect ofthe benefaction was that the levy on explained not only by historical circumstances but also by the initiative the
heads in Tenos would cease thenceforth. The imperial authorities had no communities ofthe Empire continued to have in managing even such impor-
objection to this - they were not interested in tributum capitis as such - pro- tant affairs as how to discharge their tributary obligations.
vided the community ofTenos continued to remit the fixed sum that this One arrives at the same conclusions by studying the meaning of
tax had produced. tributum in the best-known literary sources. In Taatus, for example, the
Taxation is a historical grocess, subject to changesand fluctuations
in forms ofrevenue raisingaswellasin amounts.Thereareobviousrisksin 25 Thevariableappearsto bewhetheror not the yield ofthe tax is stable. Onesuspects
making statements about it that are supposed to be valid for centuries at a thatwhena faxis newlyimposed,its yieldremainssomewhatuncertainfor severalyears
time. Thus, it may well be wrong to infer from the Tenos inscription that, (e.g., the British situation polemically described in Cassius Dio 62.3.2-3). It would be
unwise to allow substitution until one was certain of, or content with, the total that the
at any time in the past, the imperial government would have been equally tax in question produced.
26 Schwahn's account of taxation in Syria and Judaea ("Tributum, " coll. 44-7) illus-
22 See the discussion in n. 19*. trates the undesirability of talking in terms of "land" and "head" taxes while leaving out
23 Incontestingtraditionalaccountsofearlyimperialtaxation,I owemuchto Francesco of account the levies on livestock, businesses, movement of goods, etc. The total revenue
Grelle, Stipendium vel tributum": L'imposizionefondiaria nelle dottrine giuridiche del II e III fromtheprovince,inwhichthe imperial government's principal interest resided, depea-
secolo, Publicazioni della Facolta giuridica dell'Universita di Napoli 66 (Naples, 1963), dedonpersistentcollectionofthem all, andnot only oftheonesthatmayhaveproduced
whoseanalysisofthe evidencedownto Diocletianis muchmorethoroughthan mine. the largest share of the whole.*
Owing to the limited impact of his work up to now (e. g., virtually none on Ccrati, 27 Hence the judicial opinions of general import in the Digest bearing upon tributum,
Carac.annon), it is not superfluousto cover some ofthe sameterrain as he, andto insist !, e. g.. Dig. 7. 1.7.2, 7.27. 3, 25. 1. 13, 33. 2.28, 33. 2. 32. 9, 50. l5. 4-2.
on the samepoints. 28 This is usually seen from the disadvantagepusstandpoint ofcity councils' having to
24 Referredto byJones,CreekCity, p. 140withn. 86 (p. 327), alongwiththreemore make up the difference between the total owedto the government andthat yielded by
casesinwhichcommunitiesmadeuptheirquotasoftributefrom sourcesoflocalrevenue taxation (e. g.. Dig. 50. 4. 18. 26; CJ 11. 59. 1). The other side ofthe coin has left virtually
other than taxation. Two western instances appear in Grelle, Stipendium, p. 42 n. 47, who no explicit traces. Yet it appears from the silence of the sources that councils were not
commented: I municipes sono obbligati al tribute nei confronti della loro communita, strictly regulated in how they went about assessmentandcollection. Werelocal politics
non di Roma."
expected to restrain abuses in these respects?
 12 CapU t AND COLONATE PROVINCIALTRIBUTE 13

wordmostoftenreferstothe"contributions"oftheprovincialstothecentral subjects: the Cietae rebelled against their king, Archelaus of Cappadocia,
government, without implying anything about the nature of the taxes by because"hecompelled them to makedeclarations {deferrecensus)in ourway
which these contributions were raised. The deliveries ofhorses to the armies and to bear tributes. "35 It was presumably this property declaration, on
ofGermanicus that "exhausted the Gauls" ranked as tributum: so did some- whose basis the state acquired security for tax debts, that rendered tributes
thing as humble as the ox hides that constituted the sole contribution ofthe "immortal" and occasionedBoudicca's outburst against the Romans who
Frisians, and something as complex as the various taxes, regia tributa, paid exacted taxes even from the dead. 36 But the narrower sense of tributum is
by the Cappadocians to their king and then to Rome, when Cappadocia certainly not meant in all contexts where the word is used. The ox-hides
becamea province. 29The same generic senseoitributum appears to bemeant delivered by the Frisians as tributum were not levied pursuant to a census;
in passages noting a grant to certain localities oftemporary relieffrom "con- neither weretheabusivefees[solitae] thatsoldiershadto payannually to their
tributions to Rome. When the Sardinians were allowed a five-year remis- centurions "at trihutum. "87Whenlooked at in somedetail, whatthe Britons
sionof asmuchastheypaidto theaerariumortbejiscus," thewordtributum contributed to the imperial government could be classified as frumentum,
is avoided but would not have been amiss; tributum is what was remitted to tributa, and znwncta imperii muma;sa all three could be collectively called
Apamea for five years after an earthquake, and to the Byzantines for five tributain a generalcontext or whenseenfrom the seatofgovernment.39
years after the magnitudo onerum they had been made to bear by the imperial Tributa, then, means "contributions" or, in a more specific sense,
authorities during the Thracian and Bosporan wars. 30 Since the provincial "levies raised on the basisoftaxgayers' declarations. " Canweever infer that
procwatores Caesaris were certainly responsible for seeing that the rentals of the term designates a specific tax?Thepassagessuggesting this interpretation
imperialdomainsenteredthefiscus,andsincenoprovisionisotherwisemade are those where the distinction between "ordinary" and "extraordinary
for classifyingthis sourceofincome, it appearsto followthat tributumin its taxationneedsto bemade.Speciallevies,occasionedperhapsbynearbywars,
widest sense embraced domainial returns aswell. 31In any case, tributum was had as much right to be called tributa as did annually recurring charges;40
not confined to meaning imperial or state revenue; the term was broad contributions "indicted" upon subjects were not guaranteed to be identical
enough to beapplied to thefixedpayments madeto the temple ofJerusalem every year. 41On the other hand, certain contexts presuppose that provincial
by theJewsofthe Diaspora.32 revenues were predictably stable. WhenAugustus listed the total oftributa
Only rarely does Tacitus present us with tributum in a narrower in his Breviarium, or whenthe(rf6Mtoofthe Gauls were reduced by one
sense. Whenhe does, thereference isto one or more levies upon individuals quarter, or when Vespasian increased and even doubled the tributa upon the
in proportion to their declared resources. At one point in the crisis ofA. D. 69, provinces, one is forced to posit a category ofordinary revenues that could
Vitellius forced the imperial freedmen to contribute "pro numero manci- eitherbeexpectedannuallyor bescaledup or downbya statedpercentage.42
piorum ut tributum. "33A law ofNero provided "ne censibus negotiatorum In the one passage of Tacitus where this sort of distinction is explicitly
naves adscriberentur tributumque pro Ulis penderent. "34Tacitus regards this mentioned, the contrast invoked is not between "ordinary and extra-
procedure asthecharacteristically Roman wayoflevying contributions from ordinary" but between "old" charges and "new": " [Tiberius] saw to it that
the provinces not betroubled with new burdens (onera} andthat the old ones
29 Ann. 2. 5, 4. 72, 2. 56. 30 Ann. 2. 47. 2, 12. 58. 2, 12. 62-3.
31 For direct evidence of this equivalence (rents of imperial saltus in Africa), Grelle, 35 Ann. 6. 41. 1 (my emphasis). 36 Hist. 4. 32. 5; Cassius Dio 62. 3.
Stipendium, p. 17 with n. 49; to which might be added Dig. 19. 1. 52. 37 Ann. 4. 72; Hut. 1. 46. 3.
32 Tacitus Hist. 5. 5.2, wheretributaaredistinguishedasfixedfromstipes,"coMributions" 38 Agricola 13. 1, 19.4, cf. 31. 2. Also Hist. 4. 7X. 3-4, where the muniapacis ofthe Gauls are
or voluntary supplements. equated to tributa and offi-cia.
33 Hist. 2. 94. 6. 39 Agric. 32.9; Hist. 4.26. 1.
34 Ann. 13.51.2. Attentionshouldbedrawnto theimportantcontrastmadeherebetween 40 As in the case of the Byzantines, Ann. 12. 62-3.
assessment and taxes: the inclusion ofships on the census asdistinct from the payment of 41 For this verb, whose ring is particularly familiar in the later Bmpire, Hist. 5.25.4,
tribute on the assessment.The same contrast is made in a number ofthe lawsto be con- Ann. 2.60.4.
sideredlater andis essentialto their interpretation. -^ 42 Hirt. 1.46.3; Suet.Vesp. 16.
 14 CapU t AND COLONATE
PROVINCIAL TRIBUTE 15

be levied without greed or cruelty on the part of the magistrates. "43 This For example, one annotator ofTacitus tells us that, unlike the vectigalia,
provides valuableconfirmation ofthe view that tributum wasnot onetax or which were farmed by companies ofpublicans, "the direct taxes (tributa)
two but rathera. complex of onera, for the complex of charges could be were collected by government ofEcials. "47 But taxes were not "direct"
enlarged by the imposition ofnew burdens without even calling the newtax because collected by agents ofthe state rather than by contractors. "Direct"
extraordinary. The passage farther suggests that, from the standpoint of refers rather to the taxpayers; that is what Mommsen had in mind: the taxes
the central government, tribute was classed as "ordinary" only in respect to weredirectly assesseduponindividualregistered taxpayers andcollected from
itsannualyield,andnotinrespectto thespecificonerabywhichthetotalwas them. This contention, however, is not borne out in respect to collection. A
attained.In otherwords,imperialtributumwas"ordinary"to theextentthat well-known text proves that imperial officials were not involved: "I paid
it approximatedthestipendiaoftheRepublic,whichhadconsistedina fixed, tribute for that field in PudentiIIa's name; Corvinius,the public quaestor,
annual total per province. 44 was present and it was paid to him"; the collecting agent was a local magis-
Tributum was not a tax. Singular or plural, the word denoted the trate ofthe city ofOea. 48 A passage from the Digest makes the municipal
'contributions'' of the gro^^ These con- context oftaxation even more explicit: "He who has a field in another city
tributions were suggosed to be grpgqrt^^^ - mustmakedeclaration inthatcitywherethefieldlies, forhemustdischarge
commodities to jyhateach^pj -Qyince^. coyld 45
The
pr money pay. most
thetributumagriintothatcityinwhoseterritoryitisowned. "49A community-
-

equitable and efficient procedure for determining capacity to pay was to oriented process of this kind would have been superfluous if the imperial
count the provincial population and its wealth, that is to say, to hold a government had directly taxed individual provincials. What matter if lands
census in the Roman sense ofthe word. "Burdens" were assessed on the basis lay in several cities? Provided they were in the same province, the charges
of these census records, and the total they produced, whether annually or upon them could be paid en bloc to a single collecting agency. That, in fact,
under special circumstances, was the province's "tribute. " When a locality is what we begin to find happening in the latter part ofthe fourth century. 60
suffered a disaster, such as an earthquake or a plague, the emperor often Thecentrality atanearlier dateofthelocalcommunity inregistration aswell
remitted its contributionsfor a term ofyears;the samemight be doneafter as collection is a crucial obstacle to the idea that provincial taxation in the
a period when contributions to the Empire had been soheavy as to damage early Empire was "direct. " Obviously the central government hadno inten-
the local economy. What was forgiven in these cases was not simply the tion of dealing face-to-face with individual provincials; that is what local
"land tax or the "property tax"; we should understand that the governor governments were for.
and procurator were instructed to impose no onus whatever upon the Mommsen rested his case on the evidence relating to Censuses con-
community in question. 46 ducted by imperial officials, notably in Gaul. 61 Augustus personally took a
The conclusion that tributum did not mean a tax does not dispose of census there in 25^. 0. ; Germanicus is seen "occupied in conducting a census
the allegation that Augustus instituted "direct" taxation of the provinces. ofthe Gauls in A. D. 14 and detaching two legates for the same purpose in
. -.-43 Ann. 4. 6. 4.
A.D. i6; in A.D. 61, the censusper Gallias was conducted by a board ofthree
"» 44 Grelle, Stipenditim, p. 17, pointed out that, in Cicero's usage, stipeniiium is the fixed senators. 62 Epigraphy yields a scattering ofother emissaries who engaged in
contribution (total) owed by communities, whereas tributum (other than that paid by
Roman citizens) isthe levy by which the stipendium wasshared out among the taxpayers. 47 Tacitus, LoebClassicalLibrary, rv, 88 n. I.
By Livy s time, stipendium and tributum have become interchangeable, suggesting that 48 Apuleius Apologia 101. ... - 49 Dig. 50. 15.4.2.
tributum has acquired the sense of a total.* 50 Thecrucialindicatorofthisdevelopment is CTh 11.7.12 (383),whichspecifiesthat
,. 45 Schwahn, "Tributum," col. 11. thedomusofpotentiores possessores areto paytheirtaxesto theofficiumofthelocalgovernor,
, 46 Sincetribute to Romewasprimarily a chargeon the collectivity, thereliefofferedto whereascanalsareto paythroughthecuriaandlesserpeoplethroughthedefensorcivitatis.
afflicted communities was from the export ofwealth rather than from taxation. Individual CTh 11.22. i, 2 affirms the old rule, but CTh 11.22. 3 (387) permits translatio with the
I citizens might well have been "taxed" as heavily as ever - that is, they might have permission of the Praetorian Prefect.
< contributed asmuch to the community as in ordinary years - but what they paid would 51 Against Mommsen on this point, Grelle, Stipendiiim, pp. 18-19.
.

be applied to local reconstruction instead of the tribute bill. 52 Cassius Dio 53. 22. 5; Tacitus Ann. i. 31, 2. 6, 14. 46.
 l6 CapU t AND COLONATE PROVINCIALTRIBUTE 17
census-taking activities, chiefly in the western provinces. 53 The problem is counterto sucha conception.AsAbbottandjohnsonrightly said,"... in the
that we have no information about what precisely these imperial commis- provinces the unit with whichRome dealt wasrather the community than
sioners did. We never find them portrayed receiving the declarations of the individual. In accordance with this principle the tribute was ordinarily
individuals; neither can we conjecture that they were accompanied by large paid, not by the homo stipendiarius, but by the civitas stipendiaria. The pro-
staffs of underlings who would have done so in their place. There were vincial municipality therefore was made responsible for the payment of a
certainly other things for imperial commissioners to do. The compilation of certain amount ... 67 Decentralization of taxation into the hands of the
property declarations was a difficult, delicate, and unpopular business; when municipalities was the essence ofthe reform. 58Appian makes this clear in his
the Empire wasalready centuries old, Lactantius still looked upon the process version ofan address by Mark Antony to the assembly ofthe cities ofAsia:
with horror and revulsion. 54 In lands that had been orgarnz^ 'Whenthe publicans, who farmed these collections ... wronged you ...
lines, taking a census p^ Gaius Caesar remitted to you one-third ofwhat you had paid and ... turned
concept and law of real property. Close supervision was indispensable if over to you the collection ofthe taxes from the cultivators ofthe soil ... "69
actual registry was delegated to local officials, since the latter were bound to The municipalities hadnever, in fact, beenabsent from the taxation process,
be notables in their own communities and the process ofregistry gave them for the tax farmers had dealt with them. 60 Administratively, the reform
ample occasions to abuse and enhance their positions. Where property meant simply that the imperial government became its own banker and
declarations were a long-standing habit, governors could routinely handle entrusted the remaining functions of the farming companies to the cities,
the complaints and appeals that registry occasioned. There was ample reason, alongwiththosetheyhadalwaysexercised. Onemight ventureto addthat
however, to dispatch commissioners ofhigh rank and authority from Rome no measure wasbetter designed to promote the economic recovery ofplun-
to supervise censuses in provinces that were alien or new to the ways ofthe deredprovincesthantherestorationtotheirinhabitantsoftheresponsibilities,
city-state and its ideology. 65Sucha view oftheactivity ofthese officials offers and the profits, oftax assessment and collection. 61
a more satisfactory explanation of the comparative rarity and the regional Theutilization oflocal communities astheintermediaries inprovincial
distribution of the evidence than does the contention of direct imperial taxation was notjust a reform but rather the long-term policy ofthe early
assessment ofindividualprovincials. Empire. Egypt is the land where its implementation is most conspicuous.
Everyone agrees that the onset of the Principate improved the con- There, as almost nowhere else, Rome came into possession of a complex
dition ofthe subject populations, since the early emperors put an end to the bureaucratic machine for directly levying a multiplicity of taxes upon a
nefarious activities ofthe companies that hadfarmed the provincial tributes. 56 populationthathadno municipalorganization.Overa long periodoftime,
But did the reform consist ofreplacing the publicans by imperial administra- theimperialgovernmentdeliberatelydismantledthecentralizedorganization
tors dealing directly with individual provincials? The evidence goes entirely of Egypt and replaced it with municipalities of unpaid magistrates and
liturgists. 62
53 Marquardt, Staatsverwaltung, u, 208-9. On the anecdote regarding Caligula and the
Gallic census lists, below, n. 67. 57 Abbott andjohnson, Munic. Admin., pp. 120-1.
54 Lactantius De mortibuspersecutorwn23, quoted anddiscussedbelow, ch. iv. 58 Ibid. p. 185. 59 Appian Bella civilia5.4.
55 The emperor Claudius refers to the census as "novo turn opere et inadsueto Gallis" 60 Abbott andjohnson, Munic. Admin. p. 183; for the well-documented situation in late
(FIS. A11, 284-5); and note the reaction of the Cietae (above, n. 35). In attempting to Republican Sicily, Schwahn, "Tributum, " coll. 19-21.
understand the impact ofa census, it is preferable to consider the disagreeable implications 61 That there were profits as well as responsibilities in these operations should not be
of quantification among peoples unaccustomed to thinking in these terms than to dwell overlooked. Profits, honest or corrupt, from municipal contracts are surely what lay
upon the bureaucratic fairness of the process. Paying tribute to one's conqueror was a behind the mismanagement detected by Pliny in his famous tour of Bithynia (Epist.
hardship to be collectively borne more or less patiently; but systematic property declara- IO.I7, 37> 39> 110-11), andwhat inter aliamadeit advisablefor the emperorsto institute
tions had the disturbing effect of invading privacy and laying bare the inequalities of supervision in the form of curatores*
fortune within a community. 62 A.H.M. Jones, The Cities of the Eastern Roman Empire, 2nd cd. (Oxford, 1971),
56 E. g., Jones, Greek City, p. 138. pp. 314-42.The motives ofRomevis-a-visEgyptwerecertainlynot altruistic;neverthe-
 l8 CapU t AND COLONATB PROVINCIAL TRIBUTE 19

The fiscal motive ofthis policy has had the result in modern times of The bias ofthe argument is all too obvious.Jones expects us to share his
depriving Rome of any kind of credit for implementing it. Yet the social disdain for local amusements, local amenities, investments in magnificence,
benefits oflocal administration in the taxprocess should not belightly valued. contentment, and social comfort, and to condemn them as irrational because
It meant that, while the amounts ofmoney and goods siphoned away from unproductive. It is alsotaken for grantedthat laudableplanningmust be on
the locality were dictated from the outside, the allocation of resources a wider, more grandiose scale, and especially that wealth must beconstantly
within the local economy to meet thesechargesaswell asto ensure communal reinvested in the expansion ofproduction for the all-important purpose of
well-beingrested with those best situated to act in fall knowledge oflocal economic growth. Social welfare is envisaged only in these nineteenth- and
conditions, who also lived too close to the humblest members of the twentieth-century terms.
populationto ignoretheir interests. 63Modernauthorshavebeenregrettably The ways ofClassical Antiquity were not our ways, and it has yet
blind to this: to beproved that the formula we havediscoveredsucceedsin every socio-
economic context. Even if it did, we would still be obliged as historians to
The emperors thus made free use ofthe cities as local agents for indulgeandrespectthe politicalthought ofour distantforebearsin order to
executing the business ofthe central government. They left to them, understandwhattheyweredoing.65Theremaybegovernmental"responsi-
to a greater or a less extent, the assessment and collection ofdirect bilities ofa more interesting kind" than those with which the municipalities
taxes,the buildingofroads,themanagementoftheimperialpost, the ofthe Roman Empirewerevested; but the contentment they showedwith
supply of recruits, remounts, and provisions for the army. It was Roman rule might incline us to conclude that, to them at least, these were the
perhaps unfortunate that they did not see their way to trusting the onestheyweremostinterestedin. Infact, fromthestandpointofprivateand
cities with responsibilities ofa more interesting kind ... The result was communal economy, no tasks were more important. The monuments we
that the very genuine civic patriotism which flourished under the rightlyregardastheprincipalevidenceofearlyimperialprosperityowetheir
principate was directed into futile channels, the celebration of games existence far more to the role ofthe provincials in public finance than to their
and the erection of vast and unnecessary public buildings, and the ventures into productive enterprise. 66 While they were denied the choice
healthy spirit of rivalry between the cities found vent in trivial dis- ofrefusing the demands ofRome, the freedom they had in executing the
putesovertitles andprecedenceor in competitivedisplaysofmagnifi- emperor s bidding waswideenough to allow them also to remain the masters
cence which were, owing to the heavy expense which they involved, of their economic fate.
positively harmful. 64 The somewhat paradoxical conclusion these observations lead to is
that,fromthestandpointoftheimperialgovernmentasreceiveroftaxation,
less Jones's slighting comments and consistent deprecation of municipalization seem thereallydirecttaxeswerethevectigalia,whoseproceedsenteredthetreasury
excessive. For a certain corrective, rf. A. N. Sherwin-White, The Roman Citizenship from the companies farming them, whereas the ostensibly direct tributa
(Oxford, 1939), pp. 218-20. F. Oertel, Die Niedergang der hellenistische Kultur in
Aegypten, " Neuejahrbucherfur die klassische Altertum, 50 (1920), 361-81, presented Bgypt
flowed to the emperors only after having first been processed through the
asbeingat its heightunderthe exploitationofthe Ptolemaicregimeandassinkingto its assessing and collecting machinery of the communities of the Empire.
doom under the Romans, especially the later ones. This pitflessly economic standpoint Avoidinginappropriateterminology, we might simplyrecognizethat none
leaves wholly out of account the efflorescence of Greek as well as native culture in Egypt
that followed upon its municipalization.
63 The assumption is that, in the generally oligarchic condition of local politics, city 65 For a stimulating introduction to this mentality, see K. Polanyi, "Aristotle Discovers
the Economy, "
in Trade and Markets in the Early Empi res, ed. K. Polanyi et al. (Glencoe,
governmentsweredominatedbythelocalaristocracies.Theywerefreeto continuebeing
paternalisticvis-a-vistheir fellow citizens,sincethe emperors hadneither the means nor 111., 1957), pp. 64-94. Cf. M.I. Finley, "Aristotle and Economic Analysis," Past and
the inclination to wrest this function from them. Present, no. 47 (1970), 3-25.
64 Jones, Creek City, pp. 144-5. In a.similarly negativeway,J. Huizinga, Homo liidens: 66 This statement relates to the circumstances leading to municipal construction, and
A Study ofthe Play Elementin Culture (London, 1949), pp. 177-8,envisagedmunificence not to thecompositionofindividualincomes;asa recentauthorhas put it, "from exploi-
under the Empire as having been mere play, the potlatch spirit as distinct from public tationoflandandtradewealthpoured into the city aristocracies,andfrom them, by the
spirit. Suchjudgmentsappearto arisefrom a failureto recognize,or anoutrightdenialof, traditionalchannelsofaristocraticmunificence,into thecities" (E.L. Bowie,"Greeksand
the economic utility of munificence. their Past in the SecondSophistic," PastandPresent, no. 46 [1970], 39).
 20 CapU t AND COLONATE PROVINCIAL TRIBUTE 21

ofthe imperialrevenuescamefrom directtaxesin the sensethat weusethe did the amounts locally collected for remission to the imperial government
word, namely payments by individuals to agents of the highest level of compare with those collected for local purposes? Simple questions like these,
government on the basis ofassessments compiled by them. In the provinces, however important they are for an economic history ofthe Empire, cannot
imperial agents played an intermittent role in the process ofregistry by which be answered for lack ofstatistical data ofeven the grossest kind. Weare left
levies were assessed. This role wasunquestionably important, but for political to debate such constitutional problems as the relationship between the
and legal reasons rather than fiscal ones. The interest ofthe emperors expres- aerariumand thejiscus, or to inquire into administrativedetails such as the
sed itself, not in keeping track of individual fortunes (except perhaps for origins ofthe res privata. 70 Nevertheless, narrative histories ofthe ageallow
those ofRoman senators), 67 but in ensuring that the communities operated usto concludethattheimageofprovincial taxationprojected bytheDigest,
smoothly asunits ofpublic economy. That wasthe best guarantee ofimperial and indeed by the laws until well into the fourth century, faithfully reflects
finance. The alternative of directly taxing individual subjects was not what, in general, the situation had always been. 71 The emperors made
attempt ed until the later Empire, and even then only long after Diocletian. demands upon communities and received from them; their direct contact
Naturally, a systemthatallotted suchweightyresponsibilitiesto local with the taxpayers was confined to the channels of legal appeal. From the
governments was not without its problems. The stronger and more res- administrative standpoint, the tax policy ofthe early Empire was a chapter
pected the communities became, the more the emperors had to heed their ofitsmunicipal policy - theresolve to govern indirectly, to extendandfoster
pleasthattheywerecontributingasmuchastheywereable,thatanyfurther self-governing communities as the agencies ofindirect rule, to frame what-
exactionwouldbetheir ruin. Thereply ofMarcusAureliusto hissoldiers - ever laws were necessary to make the municipalities work, and to render
that a suppleaient to their pay would be bought by the blood oftheir loved judgments of such a kind as to promote communal cohesion. Modern his-
ones in the provinces68 - testifies rather to the success of taxpayers' propa- torians, while recognizing Rome's commitment to indirect government
ganda than to any fact ofexcessive taxation. The emperors' eventual resort through theimperialmunicipalities, havefelt boundto stressthatthesystem
to debasement ofthe currency as a means ofcovering their expensesillus- went sour even ata quite early date. Pliny's mission to Bithynia, theappoint-
trates their reluctanceto respondenergeticallyto communities allegingthat ment of curatores, the laws offering incentives for curial service and pre-
they could contribute no more to imperial needs than they had been. scribing appointment to municipal office in the absence of volunteers, and
Inflationis the expedientgovernmentsresort to when, for whateverreason, much else are all taken to document the breakdown ofa system - artificial
they feel unable to tax. 69 But considerations ofthis kind, while essential to remedies for a critical disease. 72Yet, asthenext chapter will show, thegovern-
a historyofRomanpublicfinance,areirrelevantto a discussionofhow trib- ment's reliance on themunicipalities at the beginning ofthe fourth century
ute was exacted from the provinces. was as great as it had ever been.
What we can know about early imperial taxation is far less than what
we must resign ourselves not to know. Did the total of tributa mount or 70 E. g., A. H. M. Jones, The Aerarium and the Fiscus, " in Studies in Roman Government
diminish in the two centuries after Augustus? Did imperial levies tend andLaw(Oxford,1960),pp. 101-14,192-4;H. Nesselhauf,"PatrimoniumandResprivata
towardsa stableannualsumpercommunity, usuallypaidin currency?How desromischenKaisers," Historia-Augiista-Colloquium,iff6j (Bonn, 1964),pp. 73-93.
71 In this sense, Grelle, Stipendwm,p. 44.
72 Fora typicalexpressionofthisdoctrine,seeF.OertelinCambridgeAncientHistory,xil
67 The one text cited to illustrate that emperors knew the sizeofindividualprovincial (Cambridge, 1939), 258-9. He andmany others have imposed upon ambiguous evidence
fortunes is CassiusDio 59.22.3-4: so as to get some money for dicing while at Lyons, a pattern ofinterpretation according towhosetenets anysignofcompulsion andimperial
[Caligula] called for the census lists ofthe Gauls and ordered the wealthiest ofthem to be intervention testifiesto municipaldecline.Accordingly,it appearsimportant to discern
putto death." Anisolatedanecdoteofthissorthardlyaffordsa securebasisforestablishing precisely when the (hypothetical) Golden Age ofthe municipalities ended andthe decline
the preciserelationship ofthe imperial government to local censuses. set in.*
68 Cassius Dio 71. 3. 3 (ed. Boissevain, m, 251-2).
69 These observations are offered as interpretative hypotheses, not as se.rure conclusions.
The fact of monetary debasement is plain, whereas documentation is inadequate to
establishupwardanddownwardtrends in imperialtaxation.
 THE SETTING OF TAXATION 23

II tax.3 Thisispossible,but thereisanalternative. Taxsystemsarealsosubject

tochange.Byrequestingimmunityfortheirland,theclergymayhavesought
no more thanto redefinetheir long-standingprivilegeswithina systemthat
was only just reaching the point oftaxing land apart from the person who
The setting of taxation in owned it.
No one in the early decades of the fourth century seemed concerned
the early fourth century about a land tax. Their worry was nomination to munera and particularly to
the curia oftheir city.

Let all doctors and ex-doctors be free and immune from all the
munera ofcurials, from the munera ofsenators, comites, andperfectissimi,
Though more abundant than in the early Empire, the evidence about and from the services (obsequia) that are frequently enjoined upon
taxation from Diocletian onwards is scanty enough that no single aspect ofit retired imperial officials; and also from all public payments (prae-
is richlydocumentedwithina narrowspaceoftime. Ifthesubjectis broken stationes). Let them also not be called to any payment (praestatio) of
down into narrow topical compartments, one can hardly avoid utilizing goldandsilver andhorsesthatmightperchancebeimposeduponthe
lawscovering more thantwo hundred years, with all the risks ofanachronism aforesaidorders and dignities.4
thatsuchanalysisentails.Analternativeisto castone'snetwidely,to embrace
as many as possible of the laws bearing on taxation, even if their ostensible This elaborate regulation, which also extended to the sons of doctors,
relation to taxesisremote, andto study them in chronological groups. Among suggestsfreedomfi-omanykindoftaxation,sincetheimmunityfrompublic
theselaws, a specialplace belongs to exemptionsfrom public burdens. Tax praestationesmust surely havesufficedto disposeofanydisembodied"land"
privileges containin themselves the imprint ofthe systemfrom whichthey tax that might havelurked about. This is borne out in the next year: "We
liberate. Their affirmations and silences are an excellent solvent of the pre- commandthatmedia,grammatici, and other professors ofletters be immune
conceptions that pervade the existing literature. together with the property (res] that they own in their cities, and that they
The earliest law granting Christian clergymen some relief from be absolved from civic magistracies (honores). 6 They were so far protected
burdensome civic service bearsthe date 313. Many more laws of the same that a heavy fine was decreed against those bringing an abusive lawsuit
sort were issued in the years to follow, ofwhich several appear intent upon against them.
completely relieving the clergy of civic concerns.1 On the other hand, When the government wished to protect certain persons from public
nothing is heard until 360 on the question whether taxes on clerical land expenses in the early fourth century, it absolved them from civic burdens.
shouldbepaid.2 Whathadchanged?Theacceptedinterpretationisthatonly Acknowledging a message from Constantine, the proconsul of Africa took
then didclergymen go sofarasto aspireto liberationfrom the generalland note that Caedlian and the clerici acting under him were "liberated from
absolutely every munus by the indulgence ofYour Majesty."6 In his own
letter, preserved by Eusebius, Constantine had said that the derid should be
i CTh 16.2. 1 and 2, on whose date,J. Gaudemet, "La legislation religieuse de Con-
stantin," Revued'histoiredeI'Eglisede France,33 (1947), 27-8; C. Dupont, "Lesprivileges
descleressousConstantin," Revued'histoireecclesiastique,62 (1967),731-3.Themostnote- 3 Jones, LRE, p. 118. Gaudemet, "Legis. relig., " pp. 29-30, and Dupont, "Privileges,"
worthy ofthe later lawsare CTh 16.2.7 (330), 8 and 10 (343), and 14 (354). p. 739, both insisted that Constantine refrained from granting tax exemptions to the
2 CTh 16.2. 15. The answergiven to the question is that, asidefrom "ea iugaet profes- clergy. For further discussion, below, n. 8*.
sionem, quae ad ecclesiam pertinet, " there is no immunity for clerical land. For comment 4 CTh 13.3.2 (320). S CTh 13.3. 1 (321).
on this phrase, see GofFart, "From Roman Taxation to Mediaeval Seigneurie: Three 6 Anullinus to Constantine, quoted in Augustine Epist. 88, Cf. CTh 16. 2.2 (313s):
Notes, " Spewlum, 46 (1972), l7X n. 31. hi, qui clerici appellantur, ab omnibus omnino muneribus excusentur.
 24 Capt it AND COLONATB
THE SETTING OF TAXATION 2$
kept once for all entirely exempted £rom all public burdens. "7 That the booked in the registersimplicatedtheseveterans with tribute. As veterans,
statement embracedtax exemptioncan scarcely be doubted, sinceno court they themselves were taxexempt, aswastheir property, provided it wasnot
could enforce collection on public account from anyone covered by such a innexus voluminous censualibus. Where had they obtained such immunity?
privilege. 8 In October ofthe same year, clerics were being vexed contrary to Though privileges for veterans have a long history, one need go no further
their privileges "by some nominations or susceptiones demanded by public back than Constantine. In 325, the land or commercial capital he granted to
custom ; Constantineorderedthat in casesofthis sort the governor should veterans was declared perpetually immune. 12 The next year, Constantine
see to it "that another be substituted to him" to perform the duty in ordered that no veteran was to be summoned to any civil munus, public
question.9 work, or payment (conlatio) by local magistrates or imperial vectigales - "let
Personalperformance(functio)ofmunerato whichonewasnominated them enjoy perpetual quiet after their labours"; furthermore, the fisc was
within a civic context, with the possibility of naming a substitute more to be ordered by letter not to disturb veterans ifthey engaged in buying and
qualified than oneselfto carry out the task- this isthe language in whichthe selling. 13 The mesh of restrictions upon local and imperial officials was
laws speak ofcostly public burdens early in the century. Municipal burdens woven tightly enough to exclude any taxation of veterans. Again, a law
were not additional to taxation. On the contrary, they were the heart of was revoked by which cohortales discharged with exemption (yacatio) of
public finance. 10 their property had been called to civil munera and curial obligations. 14
Muchlater,in400,thegovernment hadoccasiontoissuethefollowing There was no distinct land tax waiting in the wings to trouble the persons
law for Africa: "As we have learned, veterans are in possession of lands thus privileged. Veterans were still tax exempt in 400. What had changed
obligated to the registers {census) for which they disdain to pay tribute. We by then wasonly that certain landshad becometax burdenedandveterans
therefore order that whoever is found to hold lands entered in the registers werenot entitled to liberatethem from suchcharges.
should beat once compelled to pay tributes. "11Only the possession oflands Perhaps the most comprehensive ofConstantine's grants ofprivileges
was made in 319 in favour ofthe growing corps otpalatini. After specifying
7 Eusebius Hist. Eccl. 10.7. thepersonsin question, hedecreed
8 Recallingthe measuresofConstantine,a lawof343 addressed"to the clerics" (CTh
l6.2.8) says: luxta sanctionem,quam dudum meruisseperhibemini,et vos et mancipia
vestra ... vacatione gaudebitis. " In comment, it should be pointed out that, while some that neither they nor their sons and grandsons be called to a city
juristssaythatvacatiodoesnotextendtomunerapatrimoniorum{Dig.so.f.io. pr. andh.t.6, council (curia) or to magistrades {honores) or municipal munera ...
ii; CJ 10.42), the possibility of their inclusion principali beneficio is clear. Cf. Dig. To all ofthem we grant that they not undergo the duty {cwa} oftax
50.4-l8.29,whereexceptionismadeforsoldiersandveterans;thisleavesopenthepossi-
bility that other categories or individuals would be excepted, as in CJ 7. 62.7 or CTh collector or ofthe recruit collectors who are called capitularii, nor the
12.1. 1. Thesweepingexemptiongivenby Constantine to the clergyis recalled by the service(obsequium)oftemonariiorprototypiae [againrelatedto recruit-
Syro-RomanLawbook, Legessaeculares 117 (FIRASn, 794).* ing] ... Therefore, let none ofthe sons ofthe aforesaid or their slaves
9 CTh16.2.2 (3i3s). Di^. 50.4.3. 15 (Ulpian) documentstheresponsibilityofgovernors
to oversee the fairness of local apportionment of burdens. For the terminology of acquired from their personal earnings (peculium castrense) be inserted
nominatio in a.context oftaxation (exactio annonaria), see CTh 11. 30. 12 (323). in declarationsto theregisteror besubstitutedto thenumberofthose
10 Superb illustrations ofthe "liturgical" character oftaxation areprovided byP. Beatty
Panop. i (A.D. 298): wheneverthe state neededsomething, the strategus called on the 12 CTh 7.20.3 (325s); cf. h.t.4. Also h.t. 8 (364).
president, senate, or nominators ofthe Panopolitan nome to appoint, or nominate, some 13 CTh7.20.2 (326s).Theviewofjones,LRE,p.635,whichseessoldiersasimmuneonly
person or persons to carry out the task. The nominee, who bore personal responsibility from "poll tax" (his interpretation ofcaput andcapitatio), isirreconcilable with the scale of
for the amountsthatwere invariably involved (e. g.,Dig.50.4. 1.11,"The collection of exemption laid down in theseprivileges, whoseplain intent is to absolveveterans from
tributes is anonuspatrimonii"), fulfilled a more burdensome function than that of a pass all public charges. Their military service had earned them complete vacatio from further
sive agent. In the letter book ofthe strategus, the political process of nomination take- munera - a reasonable outlook for a government that did not pigeonhole "taxes" apart
precedenceoverthe amountsto becollected, whicharestatedonly when a payment is fromotherformsofpublicservice.Jones,amongothers,wasforcedtothisconclusionby
ordered.*
his reading ofcaput in the military privileges discussed below, ch. iv.
ii CTh 11. 1. 28.
U CTh 7.4. 1 (326s,).*
 26 CapU t AND COLONATB THE SETTING OF TAXATION 27

on the waiting (adcrescentes)or in the place of absentees (defi dentes),

list conlatio "indicted" upon the city by the provincial governor at the command
provided [the son or slave] is not [already] implicated in the register ofthe Praetorian Prefect. By the same expedient of beingabsolved from local
rolls. They [i. e., the palatini] are liberated from personal as well as burdens, doctors and clergymen were exempted from obligations to the
bodily munera, regardless of whether they still serve in the palace or imperial government andpalatineswereliberatedfor the emperors service.
have been given the rest they desire. 15 Imperial courts enforced nomination to local munerajust as local officials
presented recruits for the imperial army. 18 Taxation had never been an
The personal and political character of the law is striking but altogether in imperial sphere distinct from civic honoresandmunera. Becauseofthe way
keeping with the other texts discussed. By their personal service to the em- the imperial government organized its finances during the Principate, its
peror, palatini performed their civic duties and were correspondingly free practices look to us as though they were comparable to those ofmodern
from the whole range of duties, performances, and obligations forwhich states;19 only in examining local communities do we find the traditional
other citizens were liable and accountable to the registers and the courts, financial structure ofthe ancient city, in which public expenditures were met
The palarines sons had to be included in the privilege; else they would be by a combinationoflocaltaxes,returns from endowmentsandmonopolies,
called or nominated to burdensomefunctionesfor which their fathers would compulsorycontributionsbyunpaidmagistratesand"liturgists," and,incase
be financially responsible. Were palatines also free from every sort ofpublic ofdeficit,voluntarycollectionsamongrichcitizensora giftbyoneofthem.20
conlatio or praestatio, as doctors and veterans were expressly said to be? The The emperorswere familiarwith suchpracticesandknewthem to be satis-
point is open to doubt, but need not be pressed. 16 Even if they were, the factory. They hadnever created a mechanism oftaxassessment andcollection
privilege they enjoyed liberated them from what, by all appearances, paralleltothatoftheindividualcommunities.Farfromcurtailingtheactivity
constituted the most onerous part ofbeing a citizen ofthe Roman Empire. of the local res publica, their policy had been to integrate the burdens and
Therespublicaservedby a functionaryin the imperialpalace, or by a obligations ofthe greater res publica with those ofthe communities.21The
soldier on the frontiers, was identical to the res publica served by the citizen same structures that had once addressed themselves only to the petite patrie
in his city. 17 The artisan detailed with his guild by a city magistrate to clean served also to sustaina world empire.
theaqueducts, or the poor villager impressed by animperial officialto convoy
a train ofpack animals over a certain distance, performed as much ofa civic
l8 Governor's enforcement ofmunera: Dig. 50. 5. 9; h.t. 3, 15; 50. 2. 1; CJ 7. 62. 7; 10. 32. 10.
functio as the landowner contributing his considerable share to a public Presentation of recruits: Acta Maximiliani I, cd. H. Musurillo, The Acts of the Christian
Martyrs (Oxford, 1972), p. 244.
ip Abbottandjohnson,Munic. Admin.,pp. 39-54, izo-y.Sinceprovincialtributeandthe
15 CTh 6. 35. 3 (3I9S). taxes levied to raise it were largely ill cash, imperial revenues are deceptively reminiscent
16 This loophole was closed by CTh 11. 16.6 (335): palatines ofConstantinople had to of the taxation imposed by modem governments. Similarly, the imperial budget was
acknowledge pensitatio and obsequia for their capita seu iuga. This law, in which the word so structured that cash inflow and outflow largely exceededthe proportion of charges
iuga makes its first appearance in the Code (see p. 35), is an early witness to the idea of covered by muneraandvoluntary contributions (especially if onediscountsthe contribu-
a "landtax." The statement of Karayannopoulos, "DieTheoriePiganiols," p. 325, that tions of the emperors out ofdomainial returns). It is a mistake to regard such a structure
iugum, iitgatio^ caput, and capitatio all appear for the first time in the reign of Diocletian ofpublic financein Antiquity otherwisethan as anomalous and atypical.
is quite mistaken. Of these technical terms, only capitatio first occurs in a text of 20 Abbott andjohnson, pp. 138-51, 84-116 (it is symptomatic that their discussion of
Diocletian s reign. honores and munera is widely separated from that of municipal finances);Jones, Greek
17 J. Gaudemet, Civilis dans les textes juridiques du Bas-Empire, Festschrift Hans City, pp. 129-46,241-50;J. Gaudemet,InstitutionsdeI'Antiquite(Paris, 1967),pp. 516-17.
Lewald(Basel, 1953), pp. 49-51, concededthat civilis cannot be contrastedto publicus as 2i E. Kornemann, "DasProblem desUntergangsder antikenWelt," Vergangenheitund
"municipal"to "imperial";althoughthecontextoften showspublicusbeingusedto refer Gegenwart,12 (1922), 242-3, sawthe processthoughhetook a dimviewofit; for a later
to one or the other, the word itselfrefers indiscriminately to both. A good example ofthe period, and even gloomier, J. van Sickle, Diocletian and the Decline of the Roman
useful ambiguity occurs in CJ 10. 53. 1: "Cum te medicum legionis secundae adiutricis esse Municipalities,"JowrWof Roman Studies,28 (1938), 9-18. Against this traditionalview,
dicas, munera civilia, quamdiu rei publicae causa afueris, suscipere non cogeris. " The F. Grelle, Stipendium,ch. 4, hasshownthat the emperorswere at pains, duringthe third
public service the doctor performed by beingwith the army was surely as much to his century,to preventmunicipalchargesfrom becomingpassivetaxesbypreservingtothem
city as to the Empire. their characterofanactive"performance"(fmctio). I havebeengreatlyinfluencedby his
 28 Capt it AND COtONATE THE SETTING OF TAXATION 29

Accordingly, and quite unlike what our experience oftaxation leads extendthesphereofresponsibility.23Sinceclericswereexempt,accessto the
us to expect, the key questionin matters of"taxation"waswhetheror not clergy was denied in 329 not only to curials by birth but also to any persons
one belongedto the college ofcity councillors (curia) or waseligibleto be 'whose wealth was ofsuch size as to be most able to bear public charges
nominated to its expensive burdens. Who the taxpayers were was deter- {functiones}. "2i
mined not by a list arrived at on objective criteria, but by a local political The ancient polity did not believe in the equality ofcivic burdens.
processwithwhichtheemperorsinterferedonlyto a limiteddegree.By the The rich should contribute much, voluntarily or by compulsion, whilethe
turn ofthe fourth century, the key group in this respect - the one whose poor should not pay at all. It is difficult to say where the line was drawn, but
recruitment had to be sustained by imperial laws - consisted of the city we may be certain that between the category of persons bearing public
councillors. Neither infamy nor old age excused one from curial burdens; charges and those in receipt ofcivic charity there stood a substantial number
the sons of veterans were wafted into canal ranks unless theyjoined the of perfectly solvent citizens whose contribution was limited to indirect
colours in goodtime; palatineswere excusedfrom nomination;thoughno taxes.25 Long before the imperial period, in the bygone days when the
curial could bea notary, notaries could be called to the city council ifquall- politicallife ofthe Greekcitieshadbeenfree, democraticregimeshadexer-
fied; ifa curial diedwithout heirsandintestate, hisproperty descended upon cisedtheir power in sucha way as to penalizethe rich for the benefit ofthe
the college (ordo}; all councils were empowered to callJews into their ranks; generality.26The lessonwasnot lost on the oligarchiesthat normally ruled
former lessees ofimperial revenue were eligible for the council; men per- thecitiesofHellenisticandRomantimes.Theyunderstoodthatthecondition
mitted by imperialprivilegetojoin thearmed forceshadfirst to provethat oftheir pre-eminence was to continue to rechannel their wealth through the
theywereneithernominatedto thecouncilnoreligiblebybirth;mentrying city. Magistracy and "liturgy, honor andmunus, were two sides ofthe same
toavoidcurialobligationsinthecityoftheirbirthbymovingto anothercity expensive coin. One accumulated riches and otherwise exploited fellow
were to bear curial burdens in both; even provincial governors were not citizensinthewayofprivatebusiness,butinpublicmattersopen-handedness
entitled to grant a curial temporary reliefor permanent liberty from his wastherule.27Farfrom evadingtaxesor tryingto passtheburdendownto
obligations.22 Curials were tax collectors because they were, in effect, the
principaltaxpayers.Thatthiswasthenormal situationbecomesevidentfrom 23 CJ ll. 59. 1 (312-37)- One might infer from Dig. 50.4. 18.26 that Aurelian's measure
itself represented a.a extension of the burden by comparison with earlier conditions. *
the laws implying that the circle oftaxpayers was widening. Aurelianhad 24 CTh 16.2.6 (32gs); cf. CJ 11. 68. 1 (325), and above, n. 8. This drastic limitation is a
made the ordines responsible for tributes due by lands whose owners could good indicator of the different face acquired by clerical immunity in the course of
not be found; Constantine recognized that the curials might not be rich Constantine's reign.*
25 B.g. Dig. 50.5. 1.2 (quoted below, ch. m, n. 20) andCJ 11.55.1 (below, ch. m, n. 19),
enough (fdonei) for such a burden and permitted them, in case ofneed, to both of which are indicative of the limited burden of public charges weighing upon
ordinary cultivators in the third century.
argument and have sought to trace the same pattern into the fourth century, later than 26 U. von Wilamowitz-Moellendorfand B. Niese, Staat und Gesellschafl der Criechen und
Grelle himself(whosework stops beforeDiocletian) believed it to extend. Romer (Berlin-Leipzig, 1910), pp. 110-11, expressed this in tones of pathos. One cannot
22 CJ 10.32. 13 (m. 300); CTh 7.22.2 (318); 6. 35. 3 (319); 12. 1.3 (316); 5.2. I (318); help discerning anachronism in the regrets ofjones. Creek City, p. 167, that high taxation
16. 8. 3 (321); 6. 22. I (324); I2. 1. 10 (325), 12 (325), and i (326). The law of 318 {CTh oftherichwasnot practisedasa deviceforsociallevelling.Thiscomesofnot recognizing
5.2. l) is an exceptionally significant departure from previouslegislation.Diocletianfor- that liturgies were a perfectly viable form of taxation. Similarly, F. de Visscher, Les edits
badecities to claim the intestate successions ofany citizens veluti expermissu (CJ lo. io. x). d'Aiiguste decouverts a Cyreiie (Louvain, 1940), p. 88, observed that municipal liturgies
In 318, however, a tacit distinctionwas made betweenthe city andits council, and the must not beconfusedwith taxes. Perhapsnot, but taxes are in effect whatthey were.
council became the automatic heir in intestacy ofits members. Whereas Diocletian's order In an earlier age, as illustrated by W. S. Ferguson, Hellenistic Athens (London, 1911),
maintained a clear distinction between imperial property {resJisd) and municipal (res PP. 55-9, 99-l°°, democraticregimesregardedcostly magistraciesandliturgies precisely
publics), Constantines lawclearedthewayfor assertingimperialcontrol over municipal asa tax on the rich andchampionedtheir reinstatement after an oligarchicperiodwhen
lands (on the strength, e. g., ofCTh 10.8.3), sincethe property ofthe curials couldnow public functions were financed from tax receipts.*
beregardedasthe permanentmunicipalendowment.Cf. CTh 11.27. 1 (3298),wherethe 27 While recognizing and documenting the scale of generosity by the rich, Jones,
emperor appearsto be shoulderinga municipaldebt. Greek City, pp. 247-50, had some difficulty in explaining the phenomenon. Cf. Gaude-
 30 CapU t AND COLONATE

lower sociallevels, onegloried in one'sscaleofcontributions andpaidheavily Ill

for the privilege of a public memorial. 28 Nothing quite like this has ever
happened again in the history ofpublic finance; it was one of the traditions
of Antiquity that fell once for all with the end of the Roman Empire. But
in300thetraditionwasstillverymuchalive.Thenoveltyoftheagewasonly lugum and capitatlo
that the bearing of public charges had to a considerable extent been
brought under central control, out of the sphere of local politics and into
that of imperial law.

met. Institutions, p. 516, Ils [i.e., munera}ftaient rendusnccessairespar 1'insuffisancedcs

budgets locaux, mais rcpondaient aussia une certaine conception du 'service public' et
des devoirs des plus fortunes dans la cite. " This falls short ofrecognizing that munera were Among the public charges borne by the financially qualified citizens of the
simply a form oftaxation appropriateto prevalent economicconditions. late third century, one of the heaviest was the annual indictio, the levy of
28 Naturallytherewasevasion.Ulpianobserves (Dig. 50.5. 1.pr.) that, unlessexcusationes
were strictly controlled, non erunt, qui munera necessaria in rebus publicis obeant"; cf.
commoditiesimposedbytheemperorsto meetthecostsofthearmyandthe
Tertullian Apologeticiim 42. 9, quantum publico pereat et fraude et mendacio vestrarum civil service, and other expenditures. The character ofthe indictio as a tax in
professionum. The prevalent ethos, however, must be inferred from other evidence, produce was a function ofthe general economic situation, notably the col-
such as the government's unquestioned presupposition that, when expense rather than
labour was involved, fairnessconsisted in having the rich contribute first (C'Th 8. 11.1;
lapse of the currency, and the government's increasing resort to payment
ll. 15. 2); see also CTh 13. 10. 1; 11. 16. 3; and h. t. 4, with the characteristic sentence, "ea in goods as the means of maintaining its pensioners. As far as the imperial
forma servata, ut primo a potioribus, dein a mediocribus adque infimis quae sunt danda administrators were concerned, the indictio was a direct tax on cities; their
praestentur. " Antiquity had nothing to learn from us about the social blessings of the
progressive income tax. This quite recent phenomenon in modern taxation looks, by
responsibility in imposing the charge ended at the point where municipal
comparison, like a paltry and grudging step backwards to the spirit ofpublic generosity by authorities were handed the bill for the various commodities they were
the rich built into the whole structure of expensive honores and munera. See also my expected to supply in the current fiscal year. 1 The apportionment, collection,
remarks in American Historical Review, 76 (1971), 425-6. and storage of the goods in question were local responsibilities. What the
taxpayer paid was the annona, and this same term was applied to the rations
issued at the other end of the circuit to the soldiers and the other collective
or individual recipients ofgovernment salaries.2
Outside Egypt, the processes of local apportionment and collection
elude us as long as they continue to be political. 3 We learn about them when

I The process is stated in simplifiedform in order to indicatethe division ofduties. In

reality, imperialofficialscommunicatedwithmunicipalofficialsin manymorewaysthan
by presenting bills; see P. Beatty Panop. i (above, ch. n, n. 10).
2 D. van Berchem, L'annone militaire dans 1'Empire romain au me siecle, " Memoires
de la Societe nationak des antifiaires de France, 8th ser., 10 (1937), 117-202; Jones, LRE,
pp. 61, 64-5, 449, 455-9; CreekCity, pp. 143-4.*
3 The outstanding discussion of the papyri bearing on fourth-century taxation is by
Dcleage, Capitation, pp. 43-147. See also A. C. Johnson and L.C. West, Byzantine Egypt:
Economic Studies (Princeton, 1949), pp. 230-332; A. Segre, "Studies in Byzantine
Economy: lugatw and Capitatio, " Traditio, 3 (1945), IOI-26. The attention accorded here
to the papyri is small by comparison with earlier studies. None of the papyri, even the
 32 CapU t AND COLONATE
lugum AND capitatio 33
the emperors interfered to substitute objective legal forms for variable politi- cletian to restore a measure of objectivity to assessment procedures The
cal ones. Cities had considerable autonomy, but what went on in them was problem is to determinewhatthey wereandhowthey worked.
subject to the supervision of the provincial governor. His court enforced The iugum is comparatively simple to explain, since we happen to have
civic claims against the persons and property of defaulting citizens, and it a source,ratherlateindatebutreliable,thatadequatelydescribesthescheme:
entertainedappealsby citizensabusivelytreated by their peersin matters of
nomination to civic burdens. In many respects, the freedom oflocal politics The measure ofthe iugumwasdefinedin the days of Emperor Dio-
wasalreadylimitedby law,andthecloserelationshipofthegovernorto the cletian andwas made statutory: 5 iugerawhich make loplethra ofvine
municipalities ensured that irregularities in municipal life would soon were equatedto oneiugum,and20 iugeraofsowedland whichmake
occasion imperial intervention.4 40plethragivethe annonaofoneiugum;220perticaeofold olive trees
Asmattersstoodatthecloseofthethirdcentury, localapportionment give the annonaof one iugum, 450 perticae ofmountain [olive trees]
oftheannonahadacquired a novel degree offreedom asa result ofthecollapse give [theannonaof] oneiugum.Thusalsothelandthatisoflessvalue,
of money values. Currency had formerly brought a measure of objectivity asinthehills:40iugerawhichmakeSoplethragive [theannonaof] one
to local estimates ofindividual wealth. It isreasonable to suppose that, within iugum.Ifmoreovertheywereestimatedorbookedas"thirdcategory,"
a narrow confine, a city council was not disarmed in its assessment processes 6o iugeramaking 120plethragive [the annonaof] oneiugum.6
by the Suctuations of money values, particularly since the council itself
assembled the richest men of the community and very precise evaluations The Syro-Roman Lawbook goes on after this to describe the much more
were not required.8 All the same, the possibility of error and abuse had approximate modes ofassessment applied to thinly inhabited mountain and
notably widened, and if an appeal were lodged with the governor, the suit waste lands.
could be dragged on without solution. What was needed in the absence of Asit mayherebeseen,theiugumisa schedulefor transformingprop-
hard currency was an objective means of assessing wealth for purposes of erty declarationsstatedintraditionalmeasures(jjlethraor iugeraofarableand
contributing to the annona. vines, perticae of olive trees) into shares ofcontribution to the annona} To
At this point the modern debate begins. It is generally understood givea hypotheticalexample,LuciusTitiusdeclaresfifteeniugeraoffirst-class
that iugum and caput were the abstractions introduced in the reign of Dio- arable, twelve iugera of third-class arable, thirteen iugera of vineyard, and
fifty-fiveperticaeofold olive trees; by applyingthe schedule, one discovers
prefect's edict (below, n. 10), appears to me to have made a decisive contribution to thatheowesthe annonaofthreeandfour-fifthsiuga.To besure,histaxbill
clarifying the problems. It may be more useful to begin by making sense ofthe laws and
only then to return to the minutiae ofthe papyri, whoseinvaluableant's eye view of
was a variable quantity that depended on the level and form ofthe year's
taxationisoftendifficultto reconcilewiththegeneralitiesofthelaws.Foranexampleof indictio, as communicated to the local authorities. If it is assumed for the
these difficulties, below, ch. iv, n. 22.*
4 B. g., C] 4. 46. 2, anorder from thepraeses isneeded to sell property for non-payment or
arrears of tax; CJ 11.38. 1, the praeseswill assist a muntu-bearer to recover the unpaid 6 Leges saeculares 121 (FIRAS n, 795-6). Callu, Polit. monet., pp. 415-16, pushed tbe
shares of those named to the munus with him; CJ 10. 43. 1, the praeses will not suffer the authority ofthe Lawbook too far in relating to Diocletianthe tax in gold on pastures
plaintiff to be burdened unreasonably with public expenses while others are relieved. that it mentions.*
All these date from the reign of Diocletian. * 7 The traditional view is that the iugum was an "area ofcultivable land" that the schedule
5 Ulpian s description ofproperty declarations (Di^. 50. 15. 4; below, ch. w, n. 2) specifies of the Lawbook allows one to measure (Ensslin in Cambridge Ancient History, xn, 400;
that everything must be evaluated in money. No doubt the inflation made this difficult A. Segre, Studies, pp. 106-7; Deleage, Capitation, pp. 157-61). Thus, it would follow
but G. Mickwitz, CeU undWirtschaft im rSmischen Reiches deswte Jahrhunderts (Helsinki, that a vineyard of five iugera, cultivated land oftwenty iugera, or a certain number of
1932), has reminded us that persons are perfectly capable of counting in inflated values olivetrees madeup a iugum" (Abbott andjohnson, Munic. Admin.,p. 130).The tempta-
whentheneedarises.Callu,Polit.monet.,p.317,musthaveforgottentheroleofmoneyin tion to compute a concretesizefor the iugumhasobscuredits natureasa pure unit oftax
assessment when he said that, until Diocletian, "the administrative machine seems to have accountancy.The key phrasein the Lawbookis "give(s) the annonaofone iu^um." Area
been incapable of translating the elements of declaration into fiscal units that could be assuchcontinuedto be expressedin traditionalunits ofmeasure; the late Roman iugum
used for a proportional distribution [of the tax burden]." was never a survey unit imprinted on land, and no source states that it was.*
 34 CapU t AND COI. ONATB
lugwn AND capitatio 35
moment that assessment was based solely on iuga, it then follows that the Although the Syro-RomanLawbookattributes the invention ofthe
local council had to break down the government's demand into its con- iugumto Diocletian's reign, andthere is no reason to reject its testimony, the
stituent parts and divide each ofthem by the city's total ofiuga in order to worddoesnotin factoccurintheCodesuntiltheyear335;"andonehasto
determine the amount of. annonain each type ofcommodity to be paid that wait another generation to find its cognate wgatw in the sense o£professio,
yearperiugum.8 Thefinalresultwasa billpresentedto LuciusTitiusforwhat a property declaration presumably expressed in iuga. la But the prominence
he owed m respect to histhree andfour-fifths iuga. The bill washispersonal that iugum lacks at the beginning ofits course is more than made up at the
debtto the state, securedby his property.9 other end. Caput and capitatio drop out of the Codes; the Syro-Roman
Theinventionoftheiugumimplementeda reformaboutwhichweare Lawbook is silent about the caput. All the while, the iugum continues to
best informed by an edict of the prefect of Egypt, Aristius Opt atus (297) figure prominently in the laws, rightinto thereign ofjustinian. 14The iugum
We learn from him. that, as matters stood, local assessment was unfair; is more than an accountant's invention; it is also one index by which the
some taxpayers were undercharged and others overburdened. A measure was historian may trace the devolution of tax liability from persons to land.
needed that would restore some objectivity to the process. 10Sincethe iugum Before turning to this side of the problem, we should consider the more
wasanabstraction, it could beintroduced everywhere in spite ofthediversity elusiveterms ostensiblyreferringto "heads"or persons.
ofland measures in the various regions ofthe Empire. The essence ofthe From its first appearancein the early 2905 to its last in the pages of
reform wastheinstitution ofascheduleinwhoseterms property declarations, Salvian, capitatio almost invariably means "tax liability," regardless of
expressed in the ordinary local terms of survey and inventory, could be whethersuchliabilitywasincurred in respectto one'sperson or one'sland.
converted into shares ofassessment. The precise way in which the schedule Capitatio is never to be interpreted as a. tax, much less as a head or poll or
was calculated could vary aslocal conditions required. 11 capitationtax. 15The earliest datedtext speaksof "thetax liability ofa sold
field, " and equates it to "the weight or burden of performance (Junctio)"
8 A usefuldocumentforacquiringanimpressionofthisprocessisP.Oxy.1905,analysed
by Deleage, Capitation, pp. 73-4. * (293); contemporaneously, we hear of. rustid living outside city walls "who
9 His positionwouldappearto havebeenthat ofa debtorto the fisc, whose property
was pledged to the statefor hisdebt, with the state having priority in its claim over any to Bgypt is CTh 7.6. 3 (377). * On the caputin referenceto Egyptianland, below, ch. iv,
other debtor: Dig. 49. 14. 28; h. t.2l; h.t. 47. pr. ; 2. 14. 42; Fragm. de iwejisci 5 {FIRAS n, n. 38.
628); CJ 4. 9. 1 (294). The idea that property was a pledge for tax debts is clear in CTh 12 CTh 11. 16.6, discussed above, ch. n, n. 16.
11.9, De distrahendis pignoribus, quae tributorum causa tenentur. " This title was 13 CTh 8. ll. I (364). In the more general sense of property, " iugatio appears in CTh
changed in the sixth-century code, CJ 4. 46, "Si propter publicas pensitationes venditio 5. 13.2 (341); this acception is akin to professio, used in the sense of "property" in CTh
fuerit celebrata. " Inthefirstcase,thedebtwaspersonalwhileproperty wasonlya pledge; 10. 14. 2 (348s). Both laws, however, rather than referring to concrete property, may illus-
later on, the property was supposedto pay andwas soldwhen it failedto do so. trate a governmental practice ofselling or giving "deserted" tax liability; translating the
10 P. Cair. Isidor. I, in A.E.R. Boak and H.C. Youtie, The Archive of Awelius Isidorus purchased or donated assessment into actual property would have to be done subse-
(Ann Arbor, 1960), p. 29: "Our most provident Emperors ... having learned that the quently at a lower level of the administrative chain.
levies ofthe public taxes were being made capriciously so that some persons were let off 14 For some aspects of the later history of the iugum, see From Rom. Taxat. to Med.
lightly while others were overburdened, decided in the interest of their provincials to Seign., "pp. 165-75.*
root out this most evil and ruinous practice and to issue a salutary rule to which the taxes 15 The sole possible exception to this occurs in Dig. 50.4. 18.29 (Arcadius Charisius):
would have to conform. Sive autem personalium dumtaxat sive etiam civilium munerum immunitas alicui
Thus it is possible for all to know the amount levied on each aroura in accord- concedatur, neque ab annonanequeab angariisnequea veredo nequeab hospiterecipi-
ance with the character of the land, and the amount levied on each head of the rural endo neque a nave neque capitatione, exceptis militibus et veteranis, excusari possunt."
population, andthe minimum andmaximum agesofliability, from the imperial edict One might argue that, since this list consists of specific burdens (or "taxes"), capitatio
which has been published and the schedule attached thereto, to which I have prefixed for must refer to a head tax. Cf. Dig. 50.4. 18. 8, "exactores pecuniae pro capitibus" (dis-
public display the copies of this edict of mine ... The magistrates and presidents of the cussed below, ch. ry, n. 11). On the other hand, the word oddly appears at the end of the
councils ofeach city have been ordered to dispatch to eachvillage or place whatsoever a passage without a preposition, hence may be an interpolation; besides, the rationale of
copy both oftheimperial edicttogether withtheschedule andalsoofthis (edict ofmine) the passage is that the taxes in question are miwera patrimoniorum. Cerati, Carac. annon.,
as well ... "*
pp. 62-3, wastroubled by the extract andraisedthe possibility of a hay levy {capitus}
ll Not surprisingly,thepapyrialwaysspeakofarouras;theonetext relatingthewgum without liking it. The passageis difficult to build upon.*
 36 CapU t AND COLONATB
lugumANDcapitatio 37
have declared their tax liability and pay a corresponding tax-m-kind" clearly implicating landed property make it self-evident that a restrictive
{ca.290). 16Capitatio makesitslastappearanceintheCodesin430. " Not long interpretation ofthewordisinappropriate. 21Whenqualificationwasneeded,
afterwards, Salvian writes vehemently of the miseries of small farmers in the legislator supplied it, as in census humanae capitationis, "the register of
Gaul: they lose ownership (possessio) of their fields but do not lose the tax human tax liability, " or capitatio humana atque animaUum, "tax liability in
liabilityofthefields.In theflowofhissentences,capitatiojostles otherterms respect to human beings and animals. "22 Otherwise, capitatio remains a
that he regards as its synonyms: rerum tributa, vectigalia, agrorum munia.ls general word that tells us nothing about the element in respect to which
Nothingcouldmore clearly indicatethat, in hisview, capitatiois not a head liability was incurred. Close scrutiny should nevertheless be given to the
tax." laws relating to capitatio plebeia and to those where capitatio appears alongside
Most of the twenty-odd laws in which capitatio appears after 293 iugatio, or some such word.
may be disposed ofwithout commentary. In some ofthem, one might be We hear of something called capitatio plebeia from three laws issued
tempted to translate capitatio as "head tax" or as "tax liability in respect to between 343 and 368. Although this compound has often been interpreted
heads";20 but the many other uses of the term in an unquali£ed sense and as a head tax, perhaps having regional implications, 23 the laws in question
do not require that the definition be modified. For twenty-five years or so,
16 CJ 4. 49. 9 (293): Si minor a venditore ... dicebatur capitatio praediivenditi et maior
inventa sit ... Sin vero huiusmodi onus et gravamen functionis cognovisset ... " Jones,
legislators choseto specify "thetax liability ofcommoners" whenthat was
Capitatio" p. 91, stated that this lawmay refer only to the poll tax ofpersons registered the class ofthe population whose taxes were meant. The compound may
ontheestate.Theinterpretationisunconvincingm thefaceoftheplainimplicationthat have been a useful way to distinguish such liability from the glebahs collatio
praediahadcapitatio(seealsobelow,n. 36).Thereislittle basisatthisdateforspeakingof
persons registered on the estate (below, ch. v, n. 14). Karayannopoulos, "DieTheorie
of senators, the aurum coronarium levied upon curials, the collatio equorum
Piganiols, p. 326 with n. 6, claimed that whenever caput and capitatio refer to land they imposed upon honorati, and other impositions upon special groups of the
'naturallyhavethesenseoi'wgumandwgatio." Thisisincorrectandmisleading. population. 24 The chiefproblem raised by the compound is its implication
CJ ll. 55. 1 (290?),cf. CJ9.41.9. Forthetextofthislaw,andits interpretation,below, that commoners hada tax liability oftheir own different from that ofcurials;
ch. iv, n. 20.
17 CTh 11. 20. 6. Discussed in ch. iv. it derogates from the widespread modern idea ofa "general land tax" that
18 Degub.Dei5.42-3;"quodpleriquepauperculorumatquemiserorumspoliatiresculis
suis et exterminati agellis suis, cum rem amiserint, amissarum tamen rerum tributa
patiuntur; cum possessio ab his recesserit, capitatio non recedit: proprietatibus carent et 2i CJ 4.49. 9 (293)- CTA 11. 12.2 (362): those who have received immunity ofcapitatia
vectigalibus obruuntur! ... rebus eorum incubant pervasores et tributa miseri pro areexemptednotonlyfromwnonabutalsofromothercommoditiesandpaymentstothe
pervasoribus solvunt. post mortem patris nati obsequiis iuris sui agellos non habent et largitiones.CTh 1.28.2 (364): senatorsareauthorizedto establishrepresentativesin each
agrorum muniis enecantur (cd. P. Pauly, Corpus scriptorum ecclesiasticorum Latinortim, province to watch over their collective patrimonia"by also keeping track ofpaid tax
vm, 115). liability (depensaequoque capitationisrelevatione}." CTh 5. 13.4 (368): "quidquidmanci-
ip The modem equation ofcapitatio with a tax is illustrated by the titles ofseveral books piorum vel pecoris adcreverit, capitationis aut canonis augmenta non patiatur" (the
andbyinnumerablestatements;e.g., M. Bloch, Melangeshistoriques(Paris,1963), I,447; increased capital value ofthe estate isnot to result in anincrease in tax liability or fixed
Segre, Studies, p. 101. Diocletian imposed ... two new taxes, the iugatio ... and the payment). CTh 7. 13. 7 (.375)'- let each person, includiag senators, honorati, and canals, as
capitatio ; H. Bott, Die Grundzugeder diodetianischeSteuerverfassung (Darmstadt, 1928), well asplebei, pay gold in lieu ofrecruits "according to the amount ofhis tax liability
pp. 32-7, read twenty-two instances of capitatio as all meaning Grundsteuer. Yet the (promodocapitatiomssuae}." CTh11.3.5 (391):theacquirerofpropertyisatoncetomake
limitation ofcapitatio (as well as iugatio} to assessment is abundantly clear from the evi- a declaration to this effect to the census so that "tax liability" may betransferred to him.
dence. The sources yield the names ofmany taxes (annona, vestis, awumcormarium, follis CTh 7. 5.2 (404): absentees who knowingly attempt to avoid certain charges are fined
senatorius, auri lustralis collatio - to mention only the bestknown) but capitatio isnot oneof "fourtimeswhatisdemandedfor [their]tax.liability." CTh11.24.6.6 (415):thechurches
them, except in the questionable extract citedabove, n. 15. Cerati, Carac. annon., p. 428, ofAlexandria andConstantinople may retain whattheyhave "oncondition that they
statedthatcapitatiowasnot "1'impot lui-meme maissonmoded'evaluation, " addingthat bearall thefunctiones ..in accordancewiththedeclarationoftheoldtaxliability(pro
almost everyonenowacceptedthis; cf. Deleage,Capitation,p. 255,not "un impot par- antiquaecapitationisprofessione}. " CTh 11. 1. 33 (424):thechurchofThessalonica isexcepted
ticulier, maisunemethoded evaluationdela fortuneruraleen vue dela repartitionde withtheunderstandingthat it isrelievedof"onlythe amountofitstaxliability."
touteslestaxes..." Thisiscloseto thedefinitionI offer; butDeleageandCeratididnot 22 CJ ii. 52. 1 (393); CTA 11.20. 6 (430). Special attention is paid to these laws in ch. IV.
avoid confusing capitatio with a tax. 23 Forexample,Lot, Impst fancier,pp. 15-16;Jones,"Capitatio," p. 92.*
20 CJ II.55. I (290?); CTh 13. 10. 2 (313); 5. 17. 1 (332); 8. I.3 (333). Also the laws about 24 On the levies upon special groups, in general, Karayannopoulos, Finamwesen, pp.
plebeiacapitatioto be discussedshortly. 126-38, I4I-7.
 38 CapU t AND COLONATE
lugum AND capitatw 39
everyone hadto pay. Ashasbeenseen,however, capitatiowasliability, not a there was a universal land tax, from which exemption was only rarely and
tax. There was no contradiction in implying a difference in the liabilities reluctantly given. 30 Exemption is altogether plausible when one understands
ofcommoners andcurials, fortheliability ofcommoners wasinfactnarrower that tax liabilitywasincurredprimarily by personsandonly secondarilyby
thanthat ofcurials, whoweresubjectto aurumcoronariumand curialmunera land, as the security for individual functio. If, in the eyes of the authorities,
over and above their contributions to the annona, vestis, etc. the person could not "perform - owing to youth or sex or vocation - then
The earliest instance ofcapitatio plebeia occurs in the renewal ofa law he was exempt. In 331, Constantine had laid down that youths were not
of 324 according to which a man who has secured dignities by corrupt subject to nomination to curial munera or forced to undergo the obsequia
solicitation {suffragium) is reiectus in plebem, "thrust back among com- functionum until they entered their eighteenth year. Apparently, "minors to
moners. 26While the earlier lawdoesnot specifythe consequences ofbeing whom public guardianship (tutda) was due" and even children ofseven or
thus repelled, that of 343 spells them out: the men in question return to eight had been nominated. 31 The obvious reason for nominations of this
being subject to municipal munera, and they are also to bearplebeia capitatio; sort was to get at the property ofminors; public charges tended to devolve
that is to say, like other propertied men ofthe locality, they areto betaxed from persons to land. But Constantine resisted the process and so, it seems,
in proportion to their declared resources. 26Dignities did not necessarily free did Valentinian in the laws of368 and 370. The latter may in fact beregarded
onefrom taxation, butthey changedone'scategory ofliability. It wasthere- as renewals ofthe law of 331, modified to include additional categories and
fore useful to add what happened to the tax status ofpersons stripped of in general to apply to a world in which the circle oflocal taxpayers extended
dignities, "let them also bear the tax liability ofcommoners." wellbeyondthemembersofthecitycouncil.32It wouldbehastyto conclude
Thesamereading isapplicable toa lawofjulian: "Itishardly doubtful that the exemption ofwidows, wards, and virgins necessarily withdrew the^
that prototypiae and collections in respect to capitatio plebeia are munera of whole oftheir property from contribution to public charges. To the extent
curials and indeed ofthe lower kind [ofsuchmunera}. Accordingly [these that these persons - wards, in particular - required representatives at law,
munera] should be withdrawn from senatorial houses. "27 The intent of the responsibility for functiones would have devolved upon the administrators of
law was not to describe the duties ofcurials but to absolve senators from their property. 33
them. Under Julian as before, curials were the chiefcollectors of taxation, Although capitatio means"taxliability"even inthetechnical compound
their own and that ofthe local non-curials registered on the tax rolls. capitatio plebeia, three laws remain to be discussed in which its definition is
On this interpretation ofplebeia capitatio, the law in which the term more restrictive and points towards the most difficult ofthese terms, namely
last occurs prescribes immunity from taxation for widows, wards, and per- caput.
petual virgins. Widows and dedicated virgins are to be "liberated from. the
The public enemy and ours gave various persons immunity by can-
damage ofthe general tax liability"; male wards are "immune fiomfunctio
of that kind until age twenty, and women also, until each one obtains a celling the capitationes and declarations (frofessiones) ofiuga. We there-
husband. "28 The law was reissued two years later, differently worded but fore order that these privileges be altogether cancelled for everyone.
identical in substance; the same persons arenot to "acknowledge the exactly
When the beginning ofthe year is announced by the inauguration of
plebis, and the immunity ofwards continues "until those years in which
specified consuls, the requirement (necessitas) of a contribution ... is
they do not require guardians or caretakers. "29The tax exemption ofthese
persons is subject to doubt only on the assumption that, since Diocletian, kept from the humble (tenuiores), lest indeed an assessment should
be made in proportion to capitatio or iugatio and everyone should be
25 CTh6.22.1. Onthislawanditsrenewal,seemycommentin ClassicalPhilobgy,65 30 As maintained by A. Piganiol, L'impot fancier des clarissimes et des curiales au
(l97°), 146 with n. 9. Bas-Empire romain, Ecole frangaise de Rome, M'elanges d'archeohgie et d'histoire, 37
26 CTh12.1.36.Lot,ImpStfancier,p.71,understoodthislawto applyto curialswhohad (1907), I25-36.
acquireddignities;butthetextneithersaysnorimpliesthis.Beinga curialwasnota pre- 31 CTh 12. 1. 19. 32 Cf. above, pp. 28-9.
requisite for benefiting from suffragwm. 33 Cf. CJ11.62.2 (32l), onthepersonalresponsibilityofthetutor velcuratorto acquitthe
27 CTh 11.23. 2 (362). 28 CTh 13. 10.4 (368). 29 CT/I 13. 10.6 (370). charges on the imperial lands leased to their wards.
 40 CapU tAND COIONATB

squeezed as though by a legitimate [i. e., obligatory] munus... IV

Everyone who is to supply annonarian commodities in proportion
with his capitatio and sortes [is to pay them in three instalments]. 34
The three meanings of caput
In these contexts, capitatio ceases to mean "tax liability (however incurred)"
because set alongside the words professio, iugatio, and sortes, which specifi-
cally designate tax liability incurred in respect to landed property. 36 One
might interpret capitatio as "the tax liability of heads {capita), " but that is
clearly incorrect: sincethe law of340 speaks of capitationes (et professiones)
wgorum,thewordcapitatiomustrefernotto "heads"somuchasto a form of
assessment that may embrace landed property but is different from a Caput is the most difficult of the technical terms of late Roman taxation.
profession A more succinct way to say the same thing would have been to None other hasaroused greater inventiveness on the part of modern com-
use the common phrase capita seu wga, iuga vel capita: "by cancelling the mentators or occasioned them greater distress. We have nothing on the
capita or iuga, in proportion with capita or i'(^a. " The general definition of caputcomparable to the schedule for the wgum. Moreover, the texts in whicll
capitatio as "tax liability" is secure. 37It remains to discover what caput meant the term appearsdiffer widely in dateand in context. The difficulty is real
and,inthelight ofit, to establishhowcapitatiocouldhavetheall-embracing because the evidence is intractable.
sense in which we see it used from 293 to Salvian. It will be argued in the pages to follow that caput in a context of
taxation had three different but compatible meanings, and that the texts
34 CTh II.I2. I (340); 8. 11. 1(364); 11. 1. 15(367). Inaddition, CrA 11.7. 11(365) orders
thecomesreriimprivatarum to"admonishtheaccountants andprocurators [oftheresprivata] documenting the term may be satisfactorily interpreted by distinguishing
that they should strive to exceed one another in zeal to pay up whatever is demanded which senseofcaput is used in each. First, caputwasan individual "heading"
pro iugationevel capitatione," i.e., in respectto securedandunsecuredsharesofassessment in the tax registers, consisting ofthe name ofa declarant and his declaration
(== iuga vel capita). Cf. n. 36.
of taxable property. It was chiefly in this accept ion that caput was the root of
35 C&ati, Carac. annon., pp. 437-8, 441-4, was very troubled by sortes and ofFered a
complicated explanation, which seems superfluous. As he noted (p. 438), the emperor capitatio, tax liability. Secondly, caput was an "unsecured share ofassess-
Julianusedtheliteraryterm kleroi (i.e., sortes) in orderto referto iuga;sodidthedrafter ment, equivalent to the iugum in its contribution to taxation but different
ofthislaw.Cf.Cerati'sownobservations(p. 3)ontheproblemoflanguageintheCodes. from it inasmuch as the iugum was a "secured share of assessment. " The
36 In discussingthe law of340,Jones, "Capitatlo," p. 92, assertedthat it cannot mean
thecapitaoftheiuga"sinceiugacouldnotbeassessedincapita;thelinemustthereforebe caputin this sensewasthe "heading" ofsomeone who had sufficient resources
read cutting down the assessmentsof capitaand the declarations (ofland) for landed to be a contributor to taxation to the amount of one or more assessment
property. " His presupposition is disproved by the law of 293, above, n. 16. Since iuga shares, but who did not own enough real property to be worth assessing
werenot landbut "sharesofassessment," therecould becapitationesofiugajustasthere
couldbeprofessionesofcapita.TheoutstandingexampleoccursintheAutunPanegyric, through a detailed professio converted into iuga. Third, and most elusively,
to be discussedshortly. * the caputwas a "human or animal component in the formula ofassessment,"
37 C&ati,Carac.annon.,p. 431,concludedthatthetextsjust discussed"semble[nt] bieu as specified by a schedule (of which no example has survived) detailing
infirmer1 extensiondutermeuniquecapitatiocommerecouvranta lafoisladetermination
des capita et des iuga.
"

Rather, they prove that mpi'torio could be used narrowly as well as,
variations by age, sex, and category of livestock. Even if citizens' headswere
more commonly, in a broadersense.The difficulty, ifany, isto showthat capitatiocan not taxed, a schedule ofthis sort was an indispensablecomplement to the
referto landedassessmentin iuga;oncethisisdone,theconclusioneasilyfollowsthatthe schedule of iuga, for slaves and animals held a large place in individual
wordhasa restrictedsensereferringmorespecificallytoassessmentincapita.Theobserva-
tionofGrelle(abovech.n,n.21)thatthethird-centurygovernmentpreferredtoenvisage fortunes. A governmental method for converting wealth into non-monetary
taxationasactivefunctio,ratherthanpassivepraestatio,helpsto explainwhythepersonal sharesofassessmentrequiredthat a scheduleof"animate"property accom-
term capitatio was chosen as the widest expression of "tax liability." pany the schedule of "real" property. The three meanings are stated here
 42 CapU t AND COLONATE
THE THREE MEANINGS OF CORUt 43

as hypotheses that the discussion to follow will articulate and support by with pride in the public records since it was at the same time a pillar of
reference to the texts. public burdensandthusa directbenefittothecommunity. 6 Thedeclaration
The term caput is not novel to the law of the later Empire; in the on which the rich man s capitatio, "tax liability, was based normally went
concept of capitis deminutio, it had long been current to designate civil under the name ofprofessio; it might also have been called his caput, his civic
capacity, which could be changed as a result of loss of liberty (e. g., by en- capacity as viewed from the standpoint ofproperty rather than of status at
skvement), losso£civitas (e.g., by deportatio),or alterationoffamilyposition law. The same sense is given by translating caput as "heading, " in the sense
(e.g., by adoption).1 The caput of private law displays a man in motion, ofan entry in the local registers.
changing his relations to his community. By contrast, the caput of late The secondmeaningofcaput- the onemost frequently encountered
imperial tax law is at rest; it shows us the citizen participating in public in the laws - was merely a modification ofthe first. In this acception, caput
charges, as a contributor to civic needs rather than a recipient of civic wasalsoa "heading but onewhoseliability to taxation wasprecisely defined;
bounty. the term served to designate the tax-paying, or burden-bearing, capacity
As was noted before, direct financial contribution to one's munici- ofa citizen who had not made a detailed professio. Within the forms of the
pality was above all an obligation ofthe rich. Men of substance had long indictio and its collection, a taxpayer who had made a declaration supplied
been entered in the civic registers (census), whose forma, as described by the annona commensurate with his property. 6 As we saw in the hypothetical
Ulpian, survives in the Digest. 2 This declaration {professio) entailed a very caseofLucius Titius, the amount of liability to paying annonawas established
elaboratelistingofproperty:arableland,vines,olivetrees,meadows,woods, by applying to the declaration a conversion schedule laid down by the
fish ponds, landing stages, and salt pans, as the case might be, along with government, as, for example, five iugera of vine give the annona. of one
slaves, their ages, and their special skills. The extract includes a much-dis- iugum. On the other hand, a person entered in the rolls without detailed
cussed sentence. If anyone should not have declared a house-tenant or professio appeared simply as a caput, and on this basis he owed the same
tenant-farmer, he is held [personally] responsible to the registers (St quis amount of annona as someone who had declared the precise equivalent of
inquilinum vel colonum non fuerit professus, vinculis censualibus tenetur}. " In one iugum oflanded property. 7 It is wrong to infer that the person entered
other words, the owner may addto his declaration the statement that so-and- simply as a caput was without property. 8 The category was worth inventing
so, histenant, is responsible for answering to public charges in respect to the as a short cut in assessment procedure, to cover the small landowners, richer
property in question;3 if the declarant has chosen not to do so, then the in the sizeoftheir familiesthanin that oftheir acres, whosesources ofincome
responsibility falls directly on him.4
Such a declaration of property is what appeared in the tax rolls
5 The closestthe preservedevidencecomesto suchdeclarations,prominently displayed,
opposite the name ofa man ofsubstance. One's wealth could be advertised are the census inscriptions discussed in Appendix I. Not surprisingly, all the declarations
are by men of substance, except in the inscription from Hypaepa.
I W. Buckland, A Text-book of Roman Law, 2nd cd. (Cambridge, 1932), pp. 134-41. 6 His declared property was also subject to other charges; see CTh 11. 9. 2 (337); ll. 12.2
Naturally, caput could mean many other things; see TLL s.v. I draw attention to capitis (362); 13. 5. 14 (371): excusandis videlicet ... in annonario praestatione dumtaxat, ita ut
deminutiobecausethe tax caputappearsto havea genuineanalogyto caputin that other vestes adque equi ceteraeque canonicae species ab indictione eadem non negentur. The
sense.
annonaismentioned here simply becauseit wasthe heaviestchargeandthe typical term
2 Dig. 50. 15.4. for "tax, " as illustrated inter alia by C] 11. 55. 1, and the Brigetio Table (below, n. 40).
3 Cf. CTh 11. 7. 2 (319), which portrays a landowner (as it happens, a decwio) answering 7 The equal tax liability ofcaput and iugum is implicit in the texts mentioning these units
to public charges either in person or by the agency of a tenant-farmer (colonus) or a as though they were interchangeable {iuga sive, or seu, or vel, or et, capita), of which the
tributarius (presumably a settled barbarian, in line with Frag. Vat. 34; the law applies to earliest dates from 335 (above, ch. n, n. 16). The law explicitly proving the equivalence is
Britain). CTh 7.6. 3 (377)- Like iuga, capita could be expressed asfractions aswell aswhole numbers.
4 The passage is one of two in the Digest that were long regarded as documenting the 8 There could be capitationes iugorum alongside professiones of the same (as above, ch. m,
existence ofthe bound colonate prior to the fourth century: R. Clausing, The Roman Colo- n. 36). It is obvious, in any case, that owning property is not synonymous with wealth.
nate: The Theories ofIts Origins (N.Y., 1926), p. 279n. 5; A. Segre, "TheByzantine Colon- The image of a tenant richer in goods than his landlord (at least temporarily) is not alien
ate, Traditio, 5 (1947), loj.* to literature; see Seneca De beneficiis 7. 5. 2.
 44 Capl lt AND COLONATE THE THREE MEANINGS OF Capt lt 45

- for example, from the profits ofcultivating rented land - were ample to The censustakers spreadeverywhere, overturning everything: it was
qualifythemascontributorsto thetax-in-kind,but whosepropertywastoo an image ofthe tumult ofwar and ofhideous captivity. The fields
insignificant to be worth a detailed survey. were measured clod by clod, the vines andtrees were counted, animals
The result ofassessment rolls drawn up in this manner was a collection of all kinds were registered, the heads (capita) of men were noted
of professiones and capitationes adequately representing the capacity of the down;in eachcity the urbanandrural commonerswere assembled,
"active citizens" ofa district to bear public charges. By applying the con- allthepublicsquareswerefull offamiliesinflocks,everyonewasthere
version scale to the professiones, the authorities discovered how many iuga, withtheirchildrenandslaves. [Hegoesonto describethecrueltyand
with fractions, the substantial landowners were responsible for. By counting arbitrariness of the registry procedure. ] What the ancients had done
the raw capita, they established the number of additional contributors, re- by right of war to the vanquished, [Galerius] dared to do against
sponsible for one caput or several. (How someone might be assessed at more Romans and Roman subjects ... After this, money was paid for heads
thana singlecaputwillbeseena little later. ) Thecapitawerethenaddedto the (capita), and wageswere given for one's life. 10
iuga, and the resultant sum indicated the total ofassessment shares, iuga vel
capita, of the district, into which the government's variable demand for A linefromArcadiusCharisiusintheDigestalmostechoesthefinalsentence:
annona was divided. "He who receives or collects or issues annona, and collectors ofmoney for
It remains tojustify this reconstruction ofthe scheme by reference to heads,sustainthecareofa munuspersonale [i.e., supplytheirtimeandtrouble
the evidence. The key texts bearing upon capita fall into two groups. The but bear no financial charge]. u
firstcentresuponthereignofGalerius,andincludestwolaws,a passagefrom Regardless of what Lactantius might suggest regarding the personal
Lactantius, and a Gallic panegyric addressed to Constantine. The second initiative ofGalerius in this matter, the censusreferredto also extendedto the
group, whichillustratesthecaputasa "heading," consistsofa set ofmilitary westernlandsgovernedbyConstantius,thefatherofConstantine.ThePane-
privileges. Thesesourcesextendin time from thereign ofDiocletianto that gyric pronounced before Constantine in 311 or 312 on behalfofthe city of
ofConstantine, with the exception oftwo laws ofValens. After they have Autunrefersspecificallyto aneventoffiveyearsbefore,"theseverityofthe
been examined and explicated, attention will turn to the remaining evidence new registry (novi censusacerbitas)," whichsome cities ofGaul, like Reims,
about the caput, from around the turn ofthe fifth century, which leads us to could sustain with ease but which for Autun had been an economic disaster.
the third meaning ofthe term. ConstantineacknowledgedthejusticeofAutun'scomplaintbycancellingthe
Contrary to what has often been said, the sources do not ascribe to
Diocletian the institution of large-scale census activities. 9 In the pages of this by no means follows. It sufficed that the new schedules be applied to the existing
censuses.*
Lactantius, the role ofa merciless registrant oftaxpayers isplayed by Galerius, 10 Demart. pers. 23. A tradition existsofreferringthiscensusto thereignofDiocletian.
after Diocletian had retired. The passage glows with indignation: A. Piganiol, L'impot de capitation sous Ie Bas-Empire romain (Chambery, 1916), p. 20,
explicitly cited this chapter of Lactantius to document Diocletians supposed activity;
so didvan Berchem, "Annonemilit., " p. 191; Bott, Grundzuge,p. 23; andothers. More
But whatbecamea public calamityand plungedthe wholeworld in moderately, }. Moreau (Sources chretiennes 39 [Paris, 1954], p. 333 n. 2) called this a
commonmourningwasthecensusimposeduponprovincesandcities. renewal ofDiocletian's census (so also Callu, PoUt. mon'et., p. 322 n. 3). It is plain, however,
that Lactantius, who is indignant at the level and scope of Diocletians indictions (De
mart. pers. 7), doesnot ascribeto him the imposition ofa newcensus.Thisstrangeover-
9 JohnLydusDemagistratibuspopuliRoman!1.4 makesa sweepingstatementto thiseffect sight or misinterpretation ofLactantius's testimony may result from a circumstance noted
but scarcely deserves credit over the silence of Lactantius De mortibus persecutorum 7. by Moreau, pp. 7 and73, with regardto other parts ofthe tract, namelythat historical
Jones, LRE, p. 62, To provide a basisfor thenew assessments [i.e., theonesmentioned tradition was fixed (e. g., on the basisofLydus, n. 9) beforethe De mart. pers was first
in the prefects edict of 297] a series of censuses were held throughout the empire"; published (1679) andthat, as a result, its testimony has had a hard time being credited.
Johnson and West, Byzantine Egypt, p. 230, " [Diocletian] called for a complete census of ll Dig. 50. 4. 18. 8. 1 understand this "headtax" asa short-lived levy upon the urban poor
the land and people of Egypt. The logic ofthese statements is that the schedulespub- who could not supply commodities- the people subsequentlystruck from the tax rolls
lishedin 297 required newcensuses (Boak andYoutie, Arch. of Am. Isid., pp. 25-6); but (below, nn. 16-17).*
 4. 6 Capt lt AND COLONATB THREE MEANINGS
THE OF COpUt 47
recently accumulatedarrearsand grantinga reductioninassessmentofmore What had Diocletian done? He made the indictio an annual charge
thantwenty per cent. 12Therehadevenbeenquestion, Lactantiustells us, of andthusgaveItalythesamestatusasthetributaryprovinces. 18Heinstituted
registering the commoners in Rome, and this, together with a plan to shut the iugum. 19The term capitatio is first found in his reign, and it is reasonable
down the Praetorian camp, had provided the fuel for the successful usurpa- to suppose that caput in the novel sense in which we have defined it, a form
tion ofMaxentius (October 306). 13 When, at the death ofGalerius (311), of rough tax assessment, was also a novelty attributable to his time. The
MaximinusDajatried to seizetheterritory ofLicinius,"heenteredBithynia... presumptive evidence for it is in a law that has alreadybeenencountered:
and, to the greatjoy ofeveryone, cancelled the census. "14 This is best inter- 'Fromthe rural populationwhich, living outsidethe walls, hasdeclaredits
preted to mean that Maximinus cancelled the new assessment and restored taxliability andpaysa corresponding tax-in-kind, letnoonebecalledtoany
the previous one. " A little later, after Licinius recovered his lands, we find other service; neither ishe to be forced by our financial officerto undertake
him issuing a law for Lycia-Pamphylia: "Just as is the practice in the pro- the service of [escorting] imperial mules or horses."20The law disclosesan
vinces [of the diocese] ofOriens, the urban commoners are not held re- attempt to widen the body ofcontributors to expensive public burdens, by
sponsible to the registers on account oftheir tax liability. Inaccordancewith including persons from whom such sacrifices had not previously been expec-
the present command, let them rather be immune,just as the same urban ted. Members of the plebs rustica were offered an incentive for voluntarily
commoners were immune under our lord andrelative Diocletian, the retired entering themselves on the tax rolls. By doing so, they liberated themselves
Augustus. 16 The immunity enjoyed by the urban plebs under Diocletian from arbitrary but not necessarily costly requisitions oftheir time andlabour
had not been granted by special privilege. Under him, as before, they had
simply not been registered and saddled with capitatio. That had been the l8 Aurelius Victor Liber de Caesanbiis 39. 31-32; cf. Lactant. De mart. pers. 7. Eutropius
doing ofGalerius, andnow the statusquo antewasrestored, much as it had Breviarium9.23,statesvaguelythat,afterputtingdowna usurperinEgypt," [Diocletian]
been by Maximinus in Bithynia. 17 carefully laid down and organized many things that survive to our time" (i. e., the mid-
36os). The expanded Greek paraphrase ofEutropius [ca. 380) reads: "... he killed some
Egyptians and imposed as heavy taxes (phoroi) as possible on the rest. And from this
cause [i.e., therebellion], heimposed thetaxes (eisphora) ofthewholeempire, measuring
12 Panegyriques latins, ed. E. Galletier (Paris, 1952), 8(5). 5. 1, 6. 1, lo. s, Il. l, 13. 1; to be outthelandandforcing it to beregistered, andallthishasremainedinforceto thepresent
discussedshortly.Thefactthatthenewregistrywasa recentoccurrencefollowsfrom the day. Inmyview(contraBoakandYoutie, Arch.ofAurel. Isid.,p. 24), theparaphraser's
remission of five years' arrears; they had accumulated ever since the burden for Autun had additionswerebasedontheconditionsofhisowntime,withoutindependentinformation
sharplyrisen. Presumablythe reduction of capitabrought down the level of assessment about Diocletian's doings. He bears witness to the same tradition later found in more
to what hadprevailed underthe older registry (cf. below, n. 15). elaborate form in John Lydus.
13 De mart. pers. 26. 19 On the evidence ofthe Syro-Roman Lawbook, asabove, p. 33. Cf.John Lydus De
14 ttid. 36. On the date, J. Moreau, Sources chretiennes 39, pp. 398-400. mag. 1.4, whosereferenceto Diocletian's activities is best understoodto echo a statement
15 Since Moreau, op. cit., p. 399, understood this to mean that Maximinus ended taxation, such asthat ofthe Lawbook, rather than asevidence for a general census (above, n. 9).
he concludedthat the measurewas "transitory." The principle that the older assessment Littlewouldbelost ifnewevidenceprovedtheexistenceoftheassessmentwgumearlier
was the standard of fairness is clearly expressed in CTh 13. 10. 1 (313), andCassiodorus in the third century than Diocletian'sreign.
Variae 4. 38. It is one precursor of the medieval idea of the "nouvelle (et mauvaise) 20 CJll.55.i (above, ch. m,n. l6): "Nequisexrusticanaplebe,quaeextramurosposita
coQtume. For another, below, ch. v, n. 11. Segre's interpretation of sustulit census capitationem suam detulit et annonam congruam praestat, ad ullum aliud obsequium
("Studies, " p. 121) is arbitrary. devocetur neque a ratiooali nostro mularum fiscalium vel equorum ministerium subire
16 CTh 13. 10. 2 (313). cogatur. " W. Seston, Diocletien et la Tetrarchie (Paris, 1946), p. 262, had no basis for
17 Moreau, op. d(., p. 399, regarded Licinius as returning to the "errements de Dio- questioning thedateofthislaw. I fully endorsethemanner inwhichthislawwasreadby
cletien in restoring the exemption. He overlooked the likelihood (suggested by C] A.Piganiol, LacapitationdeDiocletien,"Keyue?iu(ori'^»e,176(1935),3-4,againstJones,
ll. 55-l) that theplebs urbana had never been registered before the census ofGalerius. The LRE,p. 64,whoenvisioneda distinctionbetweenheadtaxpaidin money (capitatio)and
reference isnot to a specificmeasure byDiocletian butto hisreignasa standard ofequitable land tax paid in kind {annona), asdidLot, Impotfancier, p. 15. Note also the readings of
treatment. On my interpretation, Licinius eliminated the most objectionable feature of Deleage andSeston discussed by P. Lemerle, in Revue deseludes anciennes, 50 (1948), 401
the new census, whereas Maximinus in Bithynia attained roughly the same objective by n. I. For the evidenceconfirmingthe distinctionbetweenassessment{capitatio]andtax
reinstituting the old registry. {annona}, see below, n. 41.
 48 CapU t AND COLONATE
THE THREE MEANINGS OF Caput 49
at the hands of local magistrates and imperial officials - the same sort of trates could requisition for various munera, but it also struck persons whose
requisitions to which the plebs urbana was also subject. 21The nature oftaxa- economic circumstances were such that they could contribute only money
tion, the annona in common goods readily available in the countryside, out of their miserable earnings, rather than goods that they actually pro-
rather than a tax in money, is what made possible this widening ofthe base duced; "post hoc pecuniae pro capitibus pendebantur et merces pro vita
for civicburdens.Ifrusticichosetojoin thewell-to-dointakingsharesofthe dabatur."24 Not surprisingly, the stay ofthe urban commoners on the tax
annual indictio, they acquired capita, headings, " of their own in the local rolls was ofshort duration. Otherwise, however, the great census ofGaIerius
register, with corresponding advantages. We remain in the dark, however, had enduring results. In its wake, the local tax lists of the Empire bulged
as to whether the headings in question were in the form ofdetailed profes- with unprecedented numbers ofindividual contributors to public burdens,
siones, requiring land surveys, or ofrough evaluations ofcontributive cap- enrolled together with professiones or simply as capita. The lists probably
acity, capita in the second sense ofthe term. 22 attained the widest extent they were destined ever to reach.
The course of events after Diocletian's retirement is quite clear. Whatever the situation under Diocletian may have been, the Autun
Voluntary enrolment encouraged by an incentive was replaced by simple Panegyric bears witness that the new census of306had entailed the use ofthe
compulsion: everyone wasmustered to the cities for registry and evaluated caputasa distinctivemeans ofassessment,ratherthan only asa referenceto
for contributionsto theannona,up to andincludingtheurbancommoners. 23 personal headings. In the languageofthe orator, caputis the term for a
Though the measure may not have been enforced everywhere with the share of assessment by contributors to the annona^ Gaul was where this
same comprehensiveness andseverity asLactantius observed in the dominions usageofcaput survived longest; we see it still in a poem of459, in which
ofGalerius, the Panegyric makes it plain that the West also experienced a Sidonius asks the emperor to grant him reliefofthree capita. 29
largeand, in somecases,excessiveextensionoftax liability. In the East,the Although the Panegyric contains by farthemost explicit evidence for
enrolment ofthe city plebs typified the objectionable nature ofthe new cen- caputin this sense, its passagesbearing on the form ofassessmentare com-
sus. Not only didtheir inclusioneliminatethe labourpool that local magis- paratively few. They make it apparent that persons apart from property
were an element in establishing the liability ofthe city to taxation. In two
21 I read devocetur as being addressed to municipal magistrates - the normal agents for parallelsentencestheorator explainsthatthesevereeffectsofthenewcensus
imposing obsequia - by contrast with the imperial official mentioned next. Moreover, I
interpret mularum fiscalium vel equorum ministerium" as escort service on the basis uponAutunwerenot dueto unfairnessonanyone'spart: wehavethefields
of laterevidence, CTh 11. 10. 1-2 (369-70), wheretherule isstatedthat city peopleareto thatwereassessed,andwewereestimatedby thecommonrule oftheGalli-
be used for this, not peasants dragged from field labour. It makes little difference if it were can census; again, we have the number of men and the quantity of fields
understoodinsteadto referto thesupply ofanimals,asangariae.Naturally,the law is ex-
pressed in general terms that would haveto be interpreted in practice. *
thatweredeclared.Ifthereisdoubtaboutthereality ofthesedeclaredmen,
22 The prefect s edict of 297 (above, ch. m, n. 10) might be pointed to in lieu of CJ it vanishes with the sequel. 27 Constantine, in addition to cancelling five
ll. 55. 1 to document that Diocletian instituted the caputas well as the iugum.I havenot
done so as a result ofthe puzzle presented in this respect by the papyri.* 24 De mart. pers. 23.6.
23 Lactantius specifies (De mart. pers. 23. 7-8) that only beggars were excepted and that 25 Pan'eg.lat. 8(5). ll. Thepointisbestillustratedbythesentences"septemmiliacapitum
Galerius had these rounded up and drowned. It is apparent that long-standing rules remisisti, quintam amplius partem nostromm censuum ... Remissione ista septem milium
excludedvarious categoriesfrom registryfor taxation;e. g., CTh 7.20.4 (325), "inbecilli capitum viginti quinque milibus dedisti vires, dedisti opem, dedisti salutem ... " (the gain
et debilescensibusnon dedicantur." Seemy comments onp. 29. At therisk ofanachron- instrengthbytheremainderexceedstheunitslostbyremission).Capitaaretheonlyunits
ism, I am tempted to discernthe samerules in the statement ofGregory ofTours (His- mentionedinthePanegyric.Onthedateandcircumstances,Galletier, Panegyriqueslatins,
toriamm libri 9. 30) that, in a long outdatedcensus, "viduisorfanisqueacdebilibustributi ll, 77-8.
pondusinsiderat";the inspectorsreleasedthepauperesandinfirmiandreplacedthem with 26 Carmen 13. Cf. Jones,
"

Capitatio, p. 12. Sidonius's irapM f, however, may not refer to

those whom iiisticiae condicio made tributaries (cf. Hist. 4.2, "pauperes, quos tuo debes standardized shares of assessment. *
alerehorreo, addressedto a king). It seemsasthoughfull registry ofthe populationhad 27 Paneg. 8(5). 5. 5, 6. 1, Nec tamen iuste queri poterat, cum et agros qui descripti fuerant
never been sought; the difference ca. 300 was that the line was drawn at a lower social haberemusetGallicanicensuscommuniformulateneremur,quifortunisneminipossumus
level thanin the past,whenthoseregisteredmayhavebeenveryfewby comparisonwith aequari ... Habemus enim, ut dbd et hominum numerum qui delati sunt et agrorum
the total population. The question deseryesfurther study. modum. Theformula census in the first sentence is parallel to hominumnumerum in the
 50 Capl lt AND COLONATB THE THREE MEANINGS OF CafUt 51

years' arrears, hasprovided for future reliefby eliminating 7, 000capitafrom Thefinalquestionto askishowthiscensusoperationwascarriedout.
the city's existing total of 32,000 capita. As a result, the orator says, all The fields present no problem; presumably they were declared by sub-
families are again drawn together by bonds oflove; aged parents, wives, and stantial landownersaccordingto the usual form ofprofessio.31The "heads"
adult children are respectively dearer to their offspring, husbands, and aresomethingelseagain.Theoratorsuggeststhatwhatmadelifehardinthe
parents. 28 hillyAutunnoiswasa figureofcapitaswollenbytheinclusionofagedparents,
Heads of real persons are unquestionably included in these standard- wives, andadult children. But none ofthesewere, strictly speaking, useless
sizedcapita, but the crucialpoint for an interpretation ofthe passageis that bodiesor evenunqualifiedto haveproperty oftheirown.32Asappearsinthe
agri are included as well. The capita knocked off by Constantine represent law of 340 mentioning capitationes iugorum, the caput was not unrelated to
more than one-fifth nostrorum censuum, "ofour (tax) registers, " without the property, iuga. The rusticani ofAutun had been reduced to despair by the
slightest qualification about a register of "heads" as distinct from one of newcensusbecausethefieldsno longer met thechargesofcultivation;they
"fields. " From now on, the indictio for Autun got divided on a basisof25, 000 had to go into debt and labour without return; with the cultores driven to
sharesrather than 32,ooo.29That, in effect, is the most important conclusion flight, the fields were deserted and reverted to swamp and brush. All this
to draw from the Panegyric: the capita referred to are a total ofprofessiones would beput to rights by the cancellation ofarrearsand the excision of7,000
("agrorum modus") and of capitationes ("hommum numerus"), the whole capita.Thefugitiveswouldreturnto theirpatria,ceaseto hatethesterility of
reduced by the application ofa schedule, communisformula Gallicani census, to their fields, takeheart, setto work, reopentheir houses,andrenewvowsin
uniform shares ofcontribution to the annona, collectively called capita rather the temples. 33
than wga.w Comingto termswithcaputin thePanegyricrequiresforminganidea
ofthe social and economic complexion ofAutun. Should we suppose, for
instance, that the modus agrorum was declared by the urban rich whereas the
second (in agreementwithCerati, Carac.annon., p. 528,andothers). I understandthisto
mean that the for mula contained a schedule, similar to the Syrian one for the iugum, for bare capita represented only the inhabitants of little or no property, who
converting "humannumbers" into sharesofassessment. On the other hand, the passage earneda livelihoodas tenants (co/oni) oflandowners? Suchan imageis too
should not be so restrictively construed as to limit descriptio to the fields and theformula
census to persons. There is every reason to suppose that persons as well as fields were
"assessed"(descripti), andthat, like the persons,the fieldswere siftedthroughtheformula 31 Note, however, that this form calledfor the declarationofslaves and livestock (i.e.
census. In other words, the assessment (descriptio) was conducted with the help of a perishable components) as well as real estate. This is borne out by all the evidence:
formula census that contained a schedule for agri and one for homines (cf. the edict of297, Ulpian(asabove,n. 2); Lactant. Demart.pers.23; theAsianinscriptions(seeAppendbcl);
above, ch. ill, n. 10). slaves only, CTh 6.35.3; ii.3.2; 7. 1.3 (319-49). Since sheep are counted in the Asiatic
28 Paneg. 8(5). 12. 3, "Certe et nunc liberi parentes suos cariores habentet mariti coniuges inscriptions, it follows that these perishable components were assessed as wealth, and not
non gravate tuentur et parentes adultorum non paenitet filiorum, quorum onera sibi in view oftheir "labourpower," asstill maintainedby Karayannopoulos,"DieTheorie
remissa laetantur. Ita omnium pietas olim fessa respirat et suos quemque iuvat numerare Piganiols, " p. 325.
securum, cum plures acliuvant obsequia paucorum. 32 One might infer from thePanegyricthat these persons were useless burdens to the
29 Jones, LRE, p. 1409 n. 15, insisted against the French commentators that, because headoffamily whopaidtaxes,buttheyobviouslywerenot, orneednot be.Theorator's
onlycapitaarementioned,alongwitha scheduleofpersons,thePanegyricrefersexclusively argumentwasspecious- a plausibleandcolourfulwayto express gratitude for a remi»-
to homines. This overlooks the finality with which the orator says quintam amplius sion of "heads."
partem nostrorum censuum" (ll. l). AlthoughJones, "Capitatio, " p. 90, maintained that 33 Paneg. 8(5). 14. 3, Quam multi ... quos inopia latitare per saltus aut etiam in exilium
separateregisterswerekeptfor iugaandcapita,theevidencedoesnot substantiatethepoint ire compulerat, ista remissione reliquorum m lucem exeunt, in patriam revertuntur,
unless deliberately interpreted in this sense. Only if such distinct registers were con- desinunt pristinam accusare pauperiem, desinunt odisse agrorum suorum sterilitatein,
clusively documentedcouldone entertain the possibilitythat the censuswhosetotal was resumunt animos, open praeparantur, culturam melioribus adnituntur auspiciis, revisunt
reduced by a fifth was restricted to human "heads. domos, referunt vota templis. " In short, the orator's main point is that the assessment has
30 The essentiallinks in thisargumentarethelawof340, CTh 11.12. 1, andtheinference been too heavy a charge on the total economy (land as well as people) of the Autunnois.
(above, n. 27) that theformula censuslaid downconversionsfor land aswell as persons. Teall, Age of Constantine, p. 33, was on shaky ground when he related the orators
There was a personal (or perishable) component in the professiones as well as a landed pathetic image to "the disasters of the third century. " The orator alludes to no other
component in the capitatianes:neither categorywas pure. * disaster than the novi census acerbitas.
 52 CapU t AND COLONATB THE THREE MEANINGS OP CapUt 53

simplifiedand schematica view ofthe matter. The words ofthe Panegyric The caputasa shareofassessment equivalent to thewgwnisnot quite
cannot be read in this way without being forced.34 Even the easternplebs exhausted by these explanations. In laws oflater date, and some undated
urbana pitied by Lactantius and liberated from taxation by Licinius should Asiatic inscriptions, theterm appears to havea slightly different but analogous
not be regarded as having necessarily been destitute of property. 38 For the meaning. Like the words previously discussed, caput had a future, one that
imperial and local census officials, the essence of the problem was to assess in its case tended towards its being superseded by references to land and
persons who had resources and contributive capacity but whose property dropping out of usage. But before we consider this aspect of the subject,
wasnot preciselysurveyedin themannerofmenfacedwiththeeventuality our attention should turn to the group oflaws where the senseofcaput as a
ofbequeathinga complex estate. Enteringthe tax lists wasan upward step heading" in the tax lists is most clearly marked.
in the social scale; as much is apparent from Diocletian's law about the Among the tax exemptions discussed in chapter n, there were several
plebsrustica. What mattered to both the registry ofEcial and the taxpayer was lawsofConstantine statingtheprivileges ofveteransinrather generalterms,
not so much the property in question as the declarant s personal capacity, aswell asa much later text implying that, in400, veterans were still personally
together with his family, to contribute to the annona. Land and other assets exempt from taxation, though limited in the manner in which they could
were secondary to this; their role, even in the case ofrich men, was merely enjoy their privilege. 39Four laws, one ofthem transmitted by aninscription,
to guarantee payment to the state, something to distrain in case ofdefault. specify the modalities of this exempt ion early in the fourth century and again
The smaller the property, the more illusory to supposethat anythingcould in the reign ofValens.
be obtained from it by instituting a lawsuit for recovery ofdebt. 36 By the terms of the Brigetio Table (311), "our soldiers excuse five
Unless a substantial estate was involved, the caput on the census, capita from the register {census) and from. the obligatory deliveries {praesta-
contributing annona from whatever income, completely outweighed the tiones solemnes} of annonarian payments, and they have the same [capita}
capital worth that this "heading might stand for. Precisely how the caput immune when, after completing their full term of service, they obtain an
without professio was estimated cannot be known. Apparently the formula honourable discharge. " The soldier honourably discharged after completing
census said something about it, but much was bound to be left for discussion twenty years, or discharged for wounds in battle in less than twenty years,
between censitor and declarant. From the sum of house, land, slaves, livestock, excusedtwo capitafrom the titulus annonarius, namely his own and that of
and members of the family, and by reference to the official schedule, a his wife. 40Although this inscription should not be considered apart from its
number of shares, capita, could be calculated to define the declarant's sequelintheCode,itisworthpointing outtheaffinityofitsphraseologywith
capitatio." In landsofintensivesurveying, notablyEgypt or the centuriated
parts ofAfrica, the device was superfluous and never struck roots. Elsewhere,
text rarely cited in discussions of this subject, namely the letter ofConstantine to Arius
it wasa valuable supplement to existingregistration procedures, subject to andhisfollowers (333), ed. H.-G. Opitz, Athanasius Werkeni, part I (Berlin-Leipzig,
over-enthusiasric enforcement by the officials of a harsh emperor but well 1934), 74, no- 34; also in Gelasius Kirchenyschichte3. 19.39 and41, ed. G. Loeschckeand
designed, when applied with moderation, to accommodate a broad extension M. Heinemann (Berlin, 1918), pp. 191-2; cf. H. Domes, Die SelktzeugnisKaiserKori-
in the number ofcontributors to taxation.88 stantins,Abbhandl.derAkad.derWiss.zuGottingen,philol.-hist.Klasse,3.p., 34 (1954),
lo8.*
39 Above, pp. 24-5.
34 The Panegyric does not even hint at a social differentiation ofthis kind, which modern 40 FIRASI, 457:"ut idemmilitesnostrimilitiaequidemsuaetemporequinquemcapita
historians are tempted to inject. iuxta statutum nostrum ex censu adque a praestationibus sollemnibus annonaiiae
35 Since a portion of them would soon be subjected to taxation in the form of the ami pensitarionis excusent; eademque immunia habeant adque cum completis stipendiis
lustralis coHatio, on which Karayannopoulos, Finamwesen, pp. 129-34. legitimis honestam missionem idem fuerint consecuti; sedet hii qu(i) licet post uiginti
36 Above, p. 34. stipendia adeque honestam missionem adepti fuerint, ab anaonario titulo duo kapita
37 Fordiscussion between cen^itor anddeclarant, Lactant. Demort. pers.23. 2-4. Onemight excusent, id est tam suum quam etiam uxoris suae; si quis forte ex preli uulnere causarius
assumethat, undera lessruthlessemperor,thecensitoreswouldhavebeenmoresympathetic fueritefFectus,etiamsi intra uigintistipendiaexeacausarerumsuarumuacationemfuerit
than Lactantius would have us believe those of Galerius to have been. consecutus, adbeneficiumeiusdem indulgeatiaenostiaepert(i)niat ita, ut suum et uxoris
38 A perfect illustration ofthe caput asan unsecured share ofassessment is supplied by a suae kaput excuset."
 54 CapU t AND COIONATE THE THREE MEANINGS OF CdfUt 55

that of Diocletian's law about the rusticus "who has declared his capitatio ofthree ofthese units in addition to his own (one less than in 311), and ifhe
and pays a corresponding annona" - the expressed distinction between lackedparentsanda wife,or ifanyofthesewerenotcensibusinditi,hemight
declaration and payment. 41 apply the three exemptions or less to his own slaves and land, whichwere
Constantines law of 325 contains a somewhat confusing set ofpre- presumably subject to taxation apart from the exemption of the soldier-
scriptions,distinguishingfivecategoriesofsoldiersandthreeor fourtypesof owner.Inthiswayoneattainsa definitionofcaputthatsomehistorianshave
discharges; depending on thesevariables, it grants exemptions ofone or two balkedat but that hasnevertheless hada remarkable fortune: the contributive
capitafor veterans and, ifapplicable, for their wives. 42The most noteworthy capacity ofan unadornedhuman caput becomes interchangeablewith that
clausesrelate to comitatenses, ripenses, andprotectores on actual service. As in ofa iugumofland.48Several considerationsargueagainstsucha reading.
the Brigedo Table, their exempt ion is appreciably larger than after dis- The accepted interpretation ofthese military exemptions takes caput
charge. Soldiers on active service to referto a human "head"andequatestheassessmentvalueofthatheadto
the value ofa iugum. The great difficulties that this interpretation encounters
excuse the caput ofthemselves, their father, mother, and wife, if they havehitherto beenoverlooked. To beginwith, it wouldappearthat one of
are alive and entered in the registers. Ifthey do not have one or all of the advantages offered to soldiers on active service wasa personal tax exemp-
these persons, they may excuse as much in respect to their peculium tion, immunity for his caput. Is this a reasonable inference? That soldiers on
asthey could excusein respect to thesepersons ifthey werenot want- activeduty weretax-exemptfortheir ownpersonalheadsissoself-evident a
ing, provided, however, that they excuse property {facilitates) that is privilege that no emperor could possibly have regarded himself as doing
truly their own and not someone else's property (res) by pretended them a favour by granting it. Soldiers in some modern armies pay income
ownership, on the basisofa secretagreement.43 tax, but the idea that a soldier in Antiquity should have paid a head tax goes
countertoeverythingweknowabouttheancienttheoryofobligationstothe
At first sight, the law implies a very different meaning for caput than hasbeen state.It ispointlessto objectthatthesoldierwasprivilegedformorethanone
argued in these pages.
Evenmore clearly thanthe BrigetioTable, the lawof325 allows one text in the influential theory of integrated and inseparable iugatio-capitatio developed by
Piganiol,Impotdecapitation(1916). Seston,Diocletien,p. 271,maintainedthattheBrigetio
to infer the existence ofa standard fiscal caput, composed ofinterchangeable Table implied the existence of a "caput fancier" (i.e., a caput interchangeable with a
units.It appearsasthoughthefiscal"head"stoodvery closeto thefiscalunit iugum), sincethe five capita exceeded the four possible persons in question (soldier, wife,
ofland, the iugum, since soldiers lacking "heads" offamily enjoyed a corre- father, mother); but the immunity of an extra caputcouldhavebeen appliedto a slave,
spending immunity of res, presumably land. 44Every soldier got indulgence provided the capitawere indeed human. Onthe other hand,the close oftheBrigetio Table
appears to correlate rerum siiarum vacatio with the capitaofthe discharged wounded soldier
and his wife, and the law of 325 unmistakably implies the conversion of human into in-
41 Above, n. 20. Cf. Tacit. Ann. 13. 51.2: "Ne censibus negotiatorum naves adscriberen- animate assessment.

tur tributumque pro illis penderent" (above, p. 12). The same distinction is made 45 This is virtually the crux ofthe historiographyofthe problem, for whichseeabove,
in CTh 11. 12. 2 (362); "Omnes omnino, quicumque capitationis indulgentiam immuni- Introduction, n. 3. Add:Jones, "Capitatio," pp. 93-4; M. Pallasse, La capitation et Ie
tatemque meruerunt, non solum ex annonario titulo, verum etiam ex speciebus ceteris probleme du Bas-Empire, " Revue historique deiSroitfranfais et etranger, 4th ser., 36 (1958),
atque largitionibus except! sunt immunesque erunt; neque enim praestanda dividimus"; 67-77; Faure, "Etudede la capitation," pp. 105-6;Lemerle, rev. ofDeleage,Capitation,
also in Constantine's letter of 333 (n. 38*). in Revuedesetudesanciennes,50 (1948), 399-403;andabove,n. 44.It wasdifficultto avoid
42 CTh 7.20.4. Thesemilitary detailsneednot detain us.* theconclusion that capitabecame equivalent to wga. Generally speaking, those who balked
43 Comitatenses et ripenses milites atque protectores suum caput, patris ac matris et maintained that the equivalence had eventually resulted, withthe old tributum capitis
uxoris, si tamen eos superstites habeant, omnes excusent, si censibus inditi habeantur. as its starting point. The more ambitious thesis (stimulated by Rodbertus andthe Labour
Quad si aliquam ex his personis non habuerint vel nullam habuerint, tantum pro suo Theory of Value, and comprehensively argued by Seeck) was that the equivalence was
debent peculio excusare, quantum pro iisdem, si non deessent, excusare potuissent, ita the key element ofDiocletian's reform; thus, "A capiit was the working power of a man

tamen, ut non pactione cum alteris facta simulato dominio rem alienam excusent, sed in good health" (Abbott andjohnson, Munic. Admin., p. 130). The principal element of
vere proprias facilitates." confusion, which passed unobserved, was that no one distinguished assessment from
44 Karayannopoulos, DieTheorie Piganiols, " pp. 328-9, stressed the importance ofthis taxation.
 56 CapU t AND COLONATB
THE THREE MEANINGS OF CdfUt 57
caput. Thefactthat oneofthecapitawasinvariably thesoldier's ownsuffices militaryexemptionsweresodevisedthattheimmunityofonesoldiervaried
to prove thatcaputinthesetextsmustreferto something more valuablethan capriciously from that ofthe next.
a tax exemption for the soldier's ownperson, for on no accountcould that
Thevariationsvanishifthecapitaoftheselawsareinterpretedto refer
person s head have been subject to taxation. 46
not to human heads but to "headings" in the registers - headings opposite
The soldiers' exemptions for other persons lead into further diffi- which the capitatio, "tax liability, " ofthe soldier, his father, mother, andwife
culties. It must be stressed at the outset that a caput, if human, had to be was enrolled, no matter what element of assessment such liability was
variably evaluated. Women, children, andtheagedcannot beassessedatthe incurred for. The guiding principle of exempt ion would have been the just
same figure asanadult male; even the savage census described by Lactantius and reasonable one ofallowing each soldier to benefit from an immunity as
did not do so. 47 If caput in the law of 325 really refers to a human head large or small as the resources ofhis immediate family. Some soldiers had
(variably assessed), then the son of able-bodied parents enjoyed greater rich parents, othersrich wives, others still camefrom poor familiesor were
immunity than one whose parents were aged. Even worse, how could one
destitute orphans. Thelawsdidnothing to effaceexistingdisparities ofwealth,
estimatetheamountofa soldier'specuUumthatcouldbeexcusedinplaceof a andthey necessarilyoffereda larger scaleofbenefitsto therich thanto the
wifeandparentswhomhedidnot have,whosevariablevaluecouldnot be
poor; but the same form ofexemption was allotted to all. Allowancewas
estimated? Besides, a bachelor orphan would appear to obtain greater made for soldiers without families by granting them the privilege ofliber-
personal advantages than a soldier burdened by parents, for the parents ating or withdrawing a "heading" from the tax lists by purchase, anddoing
benefitedfrom halftheson'sexemption, whilehe,presumably, wastaxable soasoften asthey hadcapitato spare; naturally, this wasa tempting privilege,
onproperty acquiredfromhispeculium. Againonthisinterpretation, thelaw open to misuse by subterfuge. 50In any case, sincerecruits werelikely tojoin
wasofgreatest benefitto landlesssonsoflandless parents. Suchfamilieswere the forces at too early an age to be entered on their local tax lists asa heading
totally exempted, whereas the immunity acquired by propertied families distinct from their fathers', they could haverealizedtheir personal exemption
through their soldier-sons left them fully liable to taxes onany land that the only by similarly applying their military earnings to the purchase ofsome
father, mother, and wife might own. Yet enough laws forbid curials to estateandwithdrawingit fromthetaxregisters.Moreover, themoreexten-
enter armed service, especially at this date, to suggest that the army wasnot sive privilege of withdrawal available to the soldiers without family com-
recruited exclusively fi-om the destitute. 48 Curials joined because of the pensated them for the fact that their fellow-soldiers with families would fall
advantages to begained, andoneisinclined to suppose thattheseadvantages heir, in part or in whole, to their parents' property. Upon discharge the
were more substantial than exemption j&om liability for personal "heads." number of exemptions shrank, but in most cases there had already been a
Finally, how could such a reading ofcaput be reconciled with the custom, reduction in the number ofheadings to be exempted. The orphan bachelor
still lively in 400 that veterans with all their property were exempt from who had "liberated" four capita with his military savings would declare a
tribute?49 Even if this problem is disregarded, it still appears as though single, consolidated (and exempt) "heading" once he wascalled upon by the
46Theexemptionofsoldiersisclearlyspelledoutin Dig.50.4.3.1 (Ulpian):"His,qui authorities to account for himself as a veteran. 61Tbejiliusfamihas had pre-
castris operam permilitiam dant, nullum municipale munus iniungipotest. Ceteriautem sumably inherited his parents' property, in part or in whole, and his wife
privati, quamvismilitumcognatisunt,legibuspatriaeet provinciaeoboediredebent."
47 As above,n. 10.Galerius'scensustakersregisteredpersonsat thehighestpossible 50 The possibility offraud hinged on the fact that the caputasa "heading"wasflexible in
assessment figure provided by the schedule, even though this meant adding years to
childrenanddeductingthemfromtheaged- anobviousindicationthat,onpaperat size.Theamountoftaxableassessment"liberated"bythesoldierwaslimitedbythesize
ofhis savings (peculitim). Ifhe dealt honestly, the government stood to lose no more than
least,a differentiationwasmade.AlsoDig.50. 15. 3 (notinglocaldifferences); CTh13. 11.2
(386). if hehadhada familyofaveragewealth. Ontheotherhand,theprivilegewouldentail
severe losses unless the soldier wasprevented from entermg into agreements with vendors
48_Crft I2. I. IO-II (325), 13 (326); 6.21.2 (326S), 13. 1 (3265); I2.I.22 (336). Gaudemet, to buy and thus "liberate" much more property than he could actually afford, on the
Constantin et les curies, " pp. 58-9, readmilitae praesidiain CTh 12.1.11 as"bureau des
gouverneurs"; this is an error. secret understandingthat hewouldsharethe profits ofthe taxfraudwith thevendor.
49 Above, ch. n, n. 11. 51 As illustrated in CTh 6. 35. 10 (380). For clarification of the conventio, CTh 12. 1. 19
(331).
 58 CapU t AND COLONATE THE THREE MEANINGS OF CapU t 59

had had a similar opportunity to inherit; the exemption of two capita since taking the oath, also excuse the tax liability of their father,
available to them sufficed to secure immunity for their wealth. mother, and wife. This withdrawalofliability entered in the lists is
Partial confirmation for such a reading ofthe laws of311 and 325 is to be made up by the enrolment ofpersons not formerly entered. 63
offered by a pair ofstatutes issued by Valens in 370 and 375, where caput
alternates with capitatio in references to the soldiers' exemption. One notes Though, in substance, these laws closely resemble the earlier pair, caput
that considerable attention is now paid to the effect upon the raxpaying only rarely appears; rather, we hear of capitatio and especially of the
community of the withdrawal of these capita (the text of the laws is not tax register [census) ofthe locality that the recruit comes from. In the light
given verbatim): ofthe common usageofcapitatio,it is hardlyconceivablethat thereference
is to a single unit of assessment composed ofa whole or fractional fiscal
From the time of taking the military oath, the recruit entered caput. The caput ofa recruit's own census is his "heading" in a tax list, con-
upon the register is absolved from the caput ofhis own registry; and, sisting ofhisnametogether with the property hehasdeclared or will declare.
after five years of active service, he obtains immunity for the tax The enlistment incentive offered to vagrants and other suchtypes isthe same
liability of his wife. This indulgence does not apply to a repudiated immunity for their "heading" in the lists, that is, chiefly, the property they
wife, who continues to endure the burden ofthe register. mighthaveor acquire.Thecaseofthewifeandparentsisevenclearer,since
Since this immunity is offered to those affixed to the register or the term capitatio is used; the soldier obtains immunity not only for their
on the waiting list, suppliers of recruits are not to present vagrants, persons but also for whatever property oftheirs causesthesarcinacensus to
veterans, fugitives, or veterans' sons. The immunity exists as an weigh upon them. Naturally, such exemption meant that the community's
incentiveto the latter categoriesto enlist voluntarily. assessment was diminished, but the government intended that there be little
The community obtains an abatement ofassessment only to the or no loss to taxation. The liability withdrawn wasto bemade up by entering
extent that the loss occasioned by the recruits cannot bereplaced in the into the tax records incensiti and adcrescentes, that is to say, persons not yet
public records from the waiting list. 62 registered and those already registered but not yet subject to contributing.
Here again, there is no reason to suppose that only "heads" ofpersons were
After detailed rules regarding the modalities ofrecruit supply, called to the active tax rolls. Rather, as many adcrescentesand incensitiwere
the law expands upon the exemptions ofsoldiers in the field army. entered as were necessary to make up for the losses in contributive capacity
After five years' active service, these soldiers, who enjoyed immunity represented by the exempted soldier andhis family. The system had acquired
a degree ofclarity such that any community could directly and mathemati-
callyrealizewhatit lostin thewayoftaxableassessmentwheneveroneofits
52 CTh7. 13.6 (370): "Sioblatusiuniorfuerit, quicensibusteneturinsertus, excotem-
pore, quo militiae sacramenta susceperit, proprii census caput excuset ac, si quinquennii
citizensbecamea soldier.Thereadyavailabilityofthisinformationwaswell
tempus fida obsequii devotione conpleverit, uxoriam quoque capitationem merito labor-
um praestetinmunem,eascilicetservandaratione,ut,quamsibiuxoremcopulaverataffectu 53 CTh 7. 13. 7. 3 (375): ... universi, qui militaria sacramenta susceperint, co anno, quo
et in priore lare derelictam memorarit, inprobata census sarcinam sustineat. Nullus vero fuerint numeris adgregati, si tamen in suscepto labore permanserint, inmunes propriis
tironem vagumautveteranumpossitofierre,cumadspontaneamsingulimilitiampropo- capitibus mox futuri sint. Conpletis vero quinque annorum stipendiis qui comitatensibus
sitaeinmunitatiscommodisinvitentur.Circacosenimlegisiubemusvalerebeneficium,qui numeris fuerit sociatus, patris quoque et matris necnonet uxoris suaecapitationem meritis
indigenas atque ipsius provinciae finibus innutritos vel adfixos censibus vel adcrescentibus
suffragantibusexcusabit.li vero, quiinripapercuneosauxiliaquefuerintconstituti,cum
suisobtulerint iuniores;nequeenim convenitillum inmunitategaudere,quivanaobla- propno capita uxorem suam tantum post quinque annos, ut dictum est, praestent in-
tione vagi atque fugitivi vel veterani filii statum futurae conventionis inviserit. Quod munem, si tamen eos censibus constiterit adtineri. Et quia publica utilitas quoque cogi-
hactenusdecernimuscustodiri,ut oblatusnumerusexadcrescentibusprimitusreparetur tanda est, ne sub hac mdulgentia insertae capitationis numerus minuatur, ex incensitis
ac, si conpensatio non potuent convenire neque ex minoribus modus, qui oblatus fuerit, adque adcrescentibus in eorum locum, qui defensi militia fuerint, alias praecipimus
quiverit reparari, itademum depublicis fascibus hi, quiexsuperfluo veniunt, eximantur." subrogari."
 60 CapU t AND COIONATE THE THREE MEANINGS OF COpUt 6l
designed to encourage the government and taxpayers alike to look with abolishes "the register ofhuman tax liability. "56Moreover, the samelaw of
favour on recruiting the army from among barbarians.54 430 initially mentioned refers to "the tax liability ofhuman beings and ani-
Caput is a complex and difficult term becausethe laws portray it in mals. "67 The only comparable element of assessment strictly limited to
three closely related but nevertheless differing senses. Up to now, in laws and persons - pure "human liability" - that earlier laws contain is confined to
other texts ofthe early fourth century, we have encountered two meanings. servi censibus adscripti, "slaves entered in the (tax) registers. " They ars first
The caputwas a "share ofassessment" equivalent to the iugum in its contribu- mentionedin 319and, accordingto lawsof327and 349, cannotbemoved
tion to the tax bill but different from the iugum in not being secured by from the province where they are registered. 68By 386, and certainly in the
property; both denoted a "share" in general, with the caput being specifically humana capitatio of 393 and 430, the reference is by no means restricted to
"unsecured" while the wgwn was "secured. " Another sense of caput was slaves. The evidence here assembled, which is supplemented by undated
closely linkedto the cognatecapitatio, "taxliability. " In thisacception,caput inscriptions fi-om the diocese ofAsiana, 69 raises several questions. Were the
meant a heading in the local assessment register, regardless ofits value and slaves declared to the registers entered as capita, and if so as whole capita?
constitution. It wasuseful in laws on exemption, since it allowed limitations And what relationship does capitatio humana vel animalium have to the caput
to besetupon exemptions without prejudice to the sizeofindividual fortunes. aswehavedefinedit for the early fourth century?
The third meaning ofcaput, which we are now coming to, begins to appear Nosoonerisa systemsetin placethanit beginsto evolve. Diocletiaa,
in the laws only in the later fourth century. Our sole contact with it hitherto in his reform law of 297, worried about unfairness on the part of local
hasbeento suspectits existencein theformula Gallicanicensusmentioned by politicians in apportioning tax liability. 60In the wake ofhis and succeeding
the Autun Panegyric. The caputin this third sense was "a perishable com- efforts, the focus ofattention tended to pass from the living apportioners and
ponent in the assessment schedule, " comparable to the "fields" we find in the payersto the objective register (census)that determinedapportionmentand
Syro-Romaa schedule for the wgum. The schedule of capita as "perishable liability. Where Diocletian had spoken ofthetax liability borne by someone
components was applied to professiones to estimate the additional assessment in respectto anestate(capitatiopraediivenditi}, Constantineakeadyspokeof
represented by slaves and livestock; it was also a device for estimating the the registered liability ofproperty [censusreicomparatae). 61As the yearswear
contributive capacity of taxpayers declaring "unsecured shares" instead of
wga.
interesting to observe that this fifth-century law wasthe key text in hisreconstruction of
A law of 430 attests that wga sive capita are equivalent to terrena seu the scheme.
animarum descriptio, "the assessment of land or souls"; it therefore justifies a 56 CTh 13. 11. 2; CJ 12. 52. 1.
gloss for caputas "the assessment ofa soul. "56Two other laws oflate date 57 Bxceptis his etc., as above, n. 55. The exception is absolute; the law never again
mentions such liability.
fall within the same order ofideas. A regulation for Pontus in 386 replaces
58 CTh 6. 35. 3 (above, ch. n, n. 15); ii.3. 2 (327); 7. 1. 3 (349). Seealso CTh ll. 1. 12 (365),
theoldnormacensa,bywhichtwowomenequalledoneman,witha newone, where the recipient of servi veluti my (the human equivalent of agri desert!) becomes
bywhichfourwomenequaltwoanda halfmen;anda lawforThracein 393 liable to fiscales pensitationes "ad integram glebae professionem, ex qua videlicet servi
videantur manere. Doesthis mean the slaves' individual plots (supposing they hadthem)
ortheentireestateonwhichtheywereregistered?Inanyevent,thelawof365isimport-
54 One recalls what Ammianus has to say (Rer. gest. 31. 4. 4) ofValens's motives for ad- ant testimony of the devolution of liability towards the land.
miningtheGothsin 376;Valensmightalsohaveregardedthisasrelieffortheprovincials 59 Discussed in Appendix I.
(31. 14.2). 6o Above, ch. m, n. 10.
55 CTh 11. 20. 6: Eorum iugorum sive capitum sive quo alio nomine nuncupantur ... 61 CTh 3. 1.2; cf. Fragmenta Vaticana 35 (the full text summarized in the Code): "ut
Exceptishis, quae in capitationehumanaatque animalium diversis qualitercumque con- omnmo qui comparat rei comparatae ius cognoscat et censum, neque liceat alicui rem
cessasunt, ita ut omnium, quae praedicto tempore atque etiam sub inclytae recordationis sine censu vel comparare vel vendere (with severe penaltiesfor infraction). Note also
avo nostro in terrena sive arumarum discribtione relevata sunt usque ad quadringentorum censusfundi comparati in CTh 11. 3. 1 (3135). It is interesting that additional legislation re-
iugorum sivecapitum quantitatem ..." lugaandcapitaareclearlydistinguishedfrom capi- quiring that land transfers be officiallyregistereddid not come until 391, CTh 11.3. 5. It
tatio humana atque animaliiim, with the latter being excluded from the application of the follows from CTh 11. 3. 1 that selling land while retaining liability hadnot only beenlegal
law.Inviewofthe greatinfluenceofGodefroyon the studyoflate Romantaxation,it is but hadoften been done (cf. Dig. 2. 14. 42). The theme continues to run through the laws
 62 CapU t AND COLONATB THE THREE MEANINGS OF CapU t 63

on, the word census leads one to many more laws about taxation than caput, mination ofthat one or several capita. Property had certainly been involved
capitatio, iugum, or iugatlo do. 62 When registers were locally compiled, re- in the calculation together with persons. 64The same applied in reverse when
sponsibility for defending the capacity ofthe locality to bear public charges the rich man s professio ofhis assets wasreduced to uniform shares (i. e., iuga),
lay chiefly with the community ; its political processes, backed by the pro- since wealth was measurable in slaves and animals as well as land. Lactantius
vincial governor if needed, moderated private inclinations towards tax bears this out. 66Thus, in the local details ofestablishing the census, the caput
evasion. Once the activities of 306 had made registration comprehensive, was not only a "heading or an "unsecured share ofassessment but also a
egalitarian,andnon-political,theimperialgovernmentacquireda largeshare "perishable" (human and animal) component in the formula ofassessment.
ofresponsibility for guaranteeingthe stability ofthe register - this in spite By contrast withthe iugum, whosesubstanceis analysedin the Syro-
of its awesome remoteness from local realities. We see the results in the laws RomanLawbook,weknownexttonothingaboutthecaputinthisacception;
of Constantine that limit the free sale ofregistered slaves and enforce the we do not even hear ofit with any clarity until the late fourth century. The
publicity of land sales, and many years later in the laws that regulate the slave adscriptus censibus cannot be sold or moved outside the province of
normaofcapita,abolishthe censushumanaecapitationis,andspeakofthe caput enrolment, but that does not mean he is worth one fall share oftax assess-
as if it were the liability of "a soul.' ment. Our only information comes from the law for Pontus in 386, which
Accordingly, the definitionofcaputto which we attain in the docu- gives usvirtually nothing to go on; it would berashto assume that the super-
ments referred to, as well as the
inscript ions, stands closer to physical reality seded norma was itself very old, since we learn from St. Basil that a new
than the abstractshareofassessmentmentionedin the documentswith which assessment affected the dominions ofValens at the opening of the 3 70S. 66
the discussionbegan. By dint ofits direct implication in enforcement, the As for capitatiohumana,it invariablyappearsin negativecontexts. Itsregister
imperial government had come to legislate at a more intimate level than in was abolished in Thrace in 393; and in addressing itself to iuga sive capita,
its lofty earlier pronouncements. 63The caputwithout professio, asit had stood understood to be terrena, the law of430 specifically exempts capitatio humana
in the register, say, of Autun, had been a share of assessment equal to a vel animalium. To be sure, negative references disclose existing things, but
iugum, but, as was earlier pointed out, we do not know the prdcesses of what is their condition? The abolition in Thrace suggests that a residual
evaluation, guided by theformula Gallicani census, that had led to the deter- portion ofthe registers was being discarded asa useless anachronism, already
moribund for years; and the disregarded capitatio of 430 bears witness to a
into the sixth century (JustinianNov. 17. 8). In the Fragmenta,Constantine'slaw on land tax system that was by now rooted exclusively in land and looked upon the
sales ostensibly dates from 313; for a discussionofthe date, Seeck, Reysten, pp. 23-4. residue of other forms of assessment merely as a nuisance. 67 As Salvian
On the point at issue here, it is obvious that the difference between capitatio and
censuswasfaint;capitatiowasno lessregisteredthancensus.Nevertheless,theargumentof
Grelle, Stipendium (above, ch. 11, n. 21), suggests the importance ofthe distinction between. 64 B. g., in the case of the rusticus who declared his capitatio (ca. 290); also capitationes tt
personal functio and passive praestatio. The two may boil down m practice to the same professiones {ugorum. (340).
thing, but it was no small matter whether the administration strove to encourage the 65 De mart. pers. 23. 2: the city squares were filled with families, complete with children
former viewpoint or, instead, turned to treating land as though it "paid." and slaves; livestock was counted along with the fields and trees. Also as above, n. 47.
62 Census as tax liability" is clearly illustrated by Symmachus Epist. 7. 126, "Res... non 66 Epist. 104 (372), which indicates that the censitores severely altered the old registry.
tam reditu ampla quam censu. Haec nisi bonorum iudicum fulciatur auxUio, publicis There was nothing to prevent the old schedulesfrom being revised. Karayannopoulos,
oneribus fessa subcumbet. " Cf. Cassiodorus Variae 7. 45. "DieTheoriePiganiols," pp. 325-6,treatedthelawof386asthoughit markeda bound-
63 Thistrait isreadilyvisiblewhenthelawsarereadin chronologicalorder.Forexample, ary line between the Diocletianic assessment of heads and later practice; though this is
several papyri (ca. 300) concern registered fruit trees and the illegality of cutting them not impossible, Basil's letter makes the inference risky. What renders useless the figures on
down (Piganiol, Capitation, " p. 4 with n. 2). It is one thing to have papyri on this sub- several ofthe Asian inscriptions (see Appendix i), aswell asthe 32, 000 ofAutun, is that we
ject andquiteanotherto encountera law, CTh 13. 11. 1 (381), dealingwithsuchadminis- lack these schedules. Regrettable though it is, we must resign ourselves to the fact that our
trative trivia. CTh 11. 28. 2 (395), excising iugera from the tax liability of Campania; ignorance of late Roman taxation is going to continue to be very great, especially in
h.t. l3 (422), doingthesamefor resprivataein Africa; andNVaknt13 (445), declaringthe matters of detail.
precise tax totals expected from two African provinces, are further instances of how some 67 Cf. NTheod. 22. 2. 12 (443); an assessment of four siliquae quas lucratlvis iugationibus
later laws attain a proximity to the soil unknown to those of the early fourth century. tantum, non humanis vel animalium censibus neque mobilibus rebus iubemus indici. *
 64 CapU t AND COLONATB
THE THREE MEANINGS OF< CapUt 65
illustrates forcontemporary Gaul,capitatlowasnowcommonly understood capita} ofsubstantial landowners; and that tenant-farmers should be pinned
asagrorummunia,the tax liability ofthe fields. 68 to the soil asa substitute for being held directly liable to taxes. " Bythe time
As the share ofassessment called a caput^ came to rest on the tax rolls it came to the end ofits useful life, aswe shall see, the caputhad called the
oftheearly fourth century, it represented a complex ofelements ofliability bound colonate into being,andtheadministration barelyremembered what
that consisted chiefly oflanded property but also took into account other the term had originally meant. 72
formsofwealth,suchasslaves,livestock, andworkingmembers ofa family.
Astimewentonandtheregistersunderwentsmallmodifications, thecap'ut 71 I donot intend to imply thatcapitatioliumanavanishedfrom theearth. Although later
began^to dissociate its hard, enduring, and inanimate elements fromitssoft, evidence from EastandWest isfragmentary, it isabundant enoughto makesucha view
mortal, andpersonal ones.Thecaput,inotherwords, slippedawayfromthe untenable.

declaring person and towards the real property declared, which could be 72 One instance of forgetfulness is discussed above, n. 67*; the principal one below,
p. 88. A casual example occurs in CTh 7.4.32 (412), where capita aie mentioned as
measuredandelaborateduponintheyearsfollowingtheoriginal,roughly assessment units, not real heads; yet, for the sakeofvariation, the legislator uses tributarios
stateddeclaration. An alternative way for capita to be assimilated to land as a synonym. Jones, Capitatio, p. 90, listed this law as an example of a tax levied on
wasforthemtoslipvoluntarilyintotbeprofessioofa morepowerfulman,by thecaputpartoftheassessment asdistinctfrommga,buttheinterpretation isunwarranted.
Capitahereweremeantasgenerallyasin the Autun Panegyric.Cf. CTh 15. 1.34,for a
the device ofa humble taxpayer's conveying his property and services to a comparableusageofiuga.
greaterlandowner. Bothprocessesareillustrated bythefollowingcontrast.
In 332 thereceiver ofa fugitive tenant-farmer washeld responsible for the
tax liability incurred by the tenant during his absence; in 361, however,
senators were assured of indemnity from disbursing fiscalia on behalf of
tenants oftheirswhohadtakenflight, provided thesenator proved heowned
nihil ex eorumdem terris, "no land oftheirs at all. "69In other words, under
Constantine persons hadtaxliability for whichthey wereresponsible before
thelocalauthorities;ifotherpersonsseducedthemfromdevotio,theliability
passedtotheseducers. Bycontrast, personsinthelastyearsofConstantius 11
had capita^ secured by their land, for whose liability they were responsible
eithertotheiremployer, ifthedeclaredlandwaspartlyhis,orintheirown
right. The stagewassetfor whatweseeexplicitly ten yearslater: tenant-
farmersweredividedintoa firstcategorythatpaiditsowntaxestothelocal
authorities and a second category the collection ofwhose taxes was under-
takenbythelandlordorhisagent;thedivisiondependedonthequestion
whetherornotthetenantownedlandinhisownright,thatistosay,whether
he had a(landed) caput ofhis own or merely formed part ofthe owner's
professio 70 From this situation the government had to take only a small
stepto decidethattheresidual, "human" taxliability oftenant-farmers was a
nuisance;thattaxesshouldbecollected from onlytheiugatio (thatis,iugavet
68 Above, ch. m, n. 18.
69 CTh 5. 17. 1; ii. 1.7. Quoted and discussed below, ch. v.
70 CTh 11. 1. 14 (37is).
 FROM COpUt TO COLONATE 67

v scheme that originally sought to assigntax liability to persons and gradually

shifted liability from elusive mortals to fixed, eternal land. Such a pattern
hasfar-reaching socialimplications. The annonaitself- a tax-in-kind charge-
able against gross local production and not limited, as was a money tax, to
From caput to colonate the proportion of production that was marketed - could be collected from
a much wider social group than the money taxes ofthe earlier Empire. A
tax-in-kind, calling most of all for common agricultural commodities like
grain, oil, and wine, madeit possibleto equatetaxableresources with pro-
duction. Thefact that liabilitywasassignedto personsmeantthat taxpayers
contributedtotheannona.regardlessofhowtheirresourceshadbeenacquired.
A landowner contributed out of his rents, but a rent-paying tenant was also
Manyofthosewhohavewritten on theiugumandcaputhavealsoaddressed qualified to contribute to the statefrom the sharethat the land yielded him. In
themselves to the colonate-not surprisingly, since several ofthe most crucial other words, without interfering with private arrangements betweenland-
documents overlap in subject matter. 1 Capltatio features prominently in the lordsandtenants,theannonamadepossiblea wideningoftheassessmentbase
law of 332 that is believed to contain the earliest reference to the colonate; toembracepersonsformerlyconfinedtothemoremenialandarbitraryforms
capitatio humana was abolished in the law of 393 that instituted the ius colo- of contribution to their communities. So matters stood for some time in the
nariumin Thrace;andthereis more.Jonesconcludedthatthetying oicoloni early fourth century. Eventually, however, the shift in assessment pro-
to the soil did not antedate Diocletian and was probably related to his cedures that drew tax liability towards land itselfbrought to prominence the
reorganization of the poll tax. "2 In closing chapter iv, I stated a different question to whom land belonged, for if rights of private property were to
hypothesis: that the caput, combining animate and landed liability, is what at continue to be respected, it was the owner who must pay the land's tax.3
length, by the removal of its animate component, called the bound colonate Thisshiftrequiredthatthelandbecultivated:theideathatthelandpaidtax
into being. A closer look at the subject is an indispensable complement to presupposed that the question whether land yielded or not was settled once
the study of terminology that hasjust been undertaken. It serves not only to for all, as a matter of public law rather than of private contract. Landpaid
illustratethesocialconsequencesoftaxlawbutalsoto confirmthedevolution to the ownerwhopaidto the state. Sucha phrase,thoughlogical onpaper,
of assessment from persons to land that constitutes the major change in hides more socio-economic reality than it reveals.4 The economic conse-
taxation during the fourth century.
The previous pages propose that, in order to be understood, the terms 3 Symmachuss famousRelatio3. 15, 18 (abouttheAltarofVictory) containsaneloquent
wgum, capitatio, and caput must be set against the backdrop of an assessment defence of private property: the same licence by virtue of which the gods are robbed
threatens all ownership. While the battle over the altar was lost, the point about private
property was not. SeeNAnth 3 (468) and cf. CTh 10. 8. 3 (326);CassiodorusI/ori(»e3. 39 and
i RothClausing,Cohnate,isa usefulsummaryoftheolderliterature.Thetheorythatthe 5.24. I leave out of account the question whether what I call "property" is not better
colonate stemmed from tax legislation was first expounded by Carl Hegel and H. A. termed seigneurie ; cf. P.W.A. Immink, "Propriety ou seigneurie? A propos des 'baux
Wallon, both of whose works were published in 1847 (Clausing, pp. 53-7). It was fully perpetuels des formules d'Angers et de Tours, " Tijdschrift voor Rechtsgeschiedenis, 29
developed by Fustel de Coulanges, "Le colonat romain," in Recherchessm quelquespro- (1961), 416-31.
blemes d'histoire (Paris, 1885), pp. 3-186, whose account closelyresemblesthat ofJones, 4 F.W. Maitland, Domesday Book and Beyond (Cambridge, 1897), p. 121, "... after all,
Colonate. " Fustel, however, alsoarguedfor significantdevelopmentstowardsthe bind- taxes must be paid by men, and not by land. " An administration that collects from "land"
ing of coloni prior to the late imperial tax laws. The view of Karayannopoulos, "Die iseventuallyboundto cut assessmentsonthe groundsthatcertainlandsdonot "pay";the
Theorie Piganiols, p. 346 with n. i, that the tax system was "neutral and ohne Einfluss productivity of land rather than men becomes the touchstone for the capacity to pay
on the colonate, is based on a syllogism, rather than on an assessment of the relevant taxes. See the cancellations of assessments mentioned above, ch. iv, D. 63. It is important
evidence, which is wholly disregarded. in readingtheselawsto realizethatthe desertionallegedisa.governmentalratherthanan
2 Colonate, " p. 4. agrarianfact; cf. CTh 11. 1. 12 (365), whereit is plainthat "deserted"landis by no means
 68 CapU t AND COLONATE FROM COpUt TO COLONATE 69

quences ofthe change to a "land tax" are too elusive to be entered into here, denied the right to evict for cause (negligence, failure to pay rent for two
but the social consequences for tenant-farmers are spelled out explicitly in the years, etc. ) or to raise rents when he changed tenants.7
laws that will now be surveyed. Thelawappearstoapplyparticularlytoa widespreadtypeoftenancy,
The combination ofa tax-in-kind with assessment based on the caput, wheretherelations oflandlord andtenant wereoflong duration, customary
a partly unsecured share of taxation, allowed the rolls of active, annona- rather than contractual. The government was protecting tenants ofthe sort
paying citizens to be extended to include the humbler elements ofthe popu- evoked long ago, and with approval, by Columella, those who were native
lation, those whose earnings derived chiefly from cultivating rented land. to theestatethey cultivated andregardedtheir tenancyashomeand patri-
Tenant-farming must have been the prevalent occupation of the "rural many. 8 Any casewhere tenancy was oflong standing, tacitly renewed from
commoners whom Diocletian invited to shedarbitrary requisitions oftheir year to year, even ifnot necessarily hereditary, came within the terms ofthe
labour byregistering liability andpaying a corresponding tax-in-kind. Those law. One may speculate that the assurance affixed rents brought about an
who did not volunteer were compelled to enter the tax rolls in the general increase in the number oftenants accepting long-term leases.
census of 306, and though the "urban commoners" were soon struck from The motives of the government in issuing the law of 325 can be
the assessment lists, the rural ones remained, relieved only by such extra- guessed at with some degree ofcertainty. It was generally concerned with
ordinary measures as that granted to Autun by Constantine. In short, the fairness in taxation, and most particularly with preventing the burden from
taxcollectors ofthe early fourth century addressed themselves not only to the beingshiftedfromtherichto thepoor. 9 A landlord facedwithhighertaxes
local landowners but also to the wider circle ofpersons who, though they onhisfortune, andwithnone ofthecompensatory prestige thatmunicipal
may have owned some land in their own right, qualified as taxpayers most munera and honores had afforded, might have been tempted to raise the rents
ofall by the profits they drew from cultivating the land ofothers. of his tenants. 10 This the government would not allow since tenants were
Three laws, of325, 332, and 361, allow us to glimpse tenant-farmers themselves taxpayers, whose earnings deserved protection. In time, the
in the period when the government looked upon them as taxpayers and principle that rents could not be raised became deeply entrenched. At the
sought to solve problems arising from this situation. There is no question yet end ofthe century, when bound tenancy had come into being, the tenant's
ofa bound colonate; theseare ordinarytenant-farmers, under the terms of right to sue because ofsuperexactio was guaranteed when all other forms of
private law. 6 What has changed is that the government has acquired an civil suit againstthe landlord weredenied.11
interest in their income and in their availability to the tax collector.
In 325, it waslaid down that therent ofa tenant could not beincreased; 7 Buckland,Text-book,pp. 498-503;W.E.Heitland, AgrimU(Cambridge, 1921),pp.
362-7.
judges were instructed to provide redress to tenants against a rent increase
8 Dererustica1.7; also,Heitland, Agricola,pp. 252-3,255, 364.
{superexactio} by their landlords and to secure them an immediate refund.6
9 Forsomeexpressionsofthis, theprefect'sedictof297 (above, ch. m, n. 10); CTh
The law mattered little if tenancies rapidly changed hands, each time ll. x6.3-4; 8. ll. i. The theme offairness,with its inevitablepresuppositionthat fairness
occasioning a specific lease. It was a breach ofcontract, in any case, to raise alwaysliesat a higherlevel intheadministrative chain,runsright through late imperial
the rent in the course ofa five-yearrental. Moreover, the landlord wasnot legislation.Itshouldnotbeunderestimatedasanimpulsetowardsadministrativechange.
10 It maynot betoofar-fetchedtorelatethislawto thereturn ofmonetary circulation in
gold in Constantine's reign; landlords might have attempted to raise the rents ofthose
uninhabited,andCJ11.59.1 (312-37),II (405),whichshowthat"deserted"wasthelegal tenants capable of producing commodities that would now fetch gold in the market.
term for landwhoseowneror registeredtaxpayercouldnot belocated. Cf. CJ 11.48.5 (366).
5 EarlierlawsintheCodesmentioningcolon!arecompletelycompatiblewithordinary ll CJII.50A(396s);II.48.23.2 (531-4);cf.C7I.5S.4 (385).Anothernoteworthyfeature
tenant-farmiug. isthetitle CJ 11.50, "la quibuscausiscolonicensitidominosaccusarepossunt," under
6 CJ il. 50. 1: Quisquis colonus plus a domino exigitur, quam ante [dare] consueverat whichthelaw0^325isentered. * Therulesurvivedvigorouslyinthebarbarianking-
et quam in anterioribus temporibus exactus est, adeat iudicem, cuius primum poterit doms; e.g., Defisco Bardmnensi (592), in ConciUos VisigSticos, ed.J. Vives (Barcelona-
haberepraesentiam, etfacinuscomprobet, utille, quiconvincitur ampliuspostulate, quam Madrid, 1963), p. 54; LegesVisigothorum12.1.2; Chlothar n, Edktum 8, in Monumenta
accipere consueverat, hoc facere in posterum prohibeatur, prius reddito quad super- Cermamaehistorica,Capituhria,I,22;L.Schiaparelli,cd.,CodicedipkmaticoLongobardo, n
exactione perpetrata noscitur extorsisse." (Rome, 1933), 18,no. 131 (758); andmuchmore. Together withthe principle that the
 70 CapU t AND COLONATB FROM CapU t TO COLONATE 71

Protecting tenants from the landlord was one thing; keeping them not his specifictenancy, for registry hasno necessaryrelation to estate. 14In
availableto the tax collector wasanother, provided for by the famous law other words, the law is not primarily concerned with landlords and their
of332: cultivators. It addresses itselfto a certain occupational category among tax-
payers, namely tenant-farmers, and introduces measures designed to force
Apud quemcumque colonus iuris alieni fuerit inventus, is non solum these free citizens to continue fulfilling their public officia in their places of
eundem origini suae restituat, verum super eodem capitationem registry. 16
temporis agnoscat. Ipsos etiam colonos, qui fugam meditantur, in Two passagesin the law, especiallywheninterpreted with the aid ot
servilem condicionem ferro ligari conveniet, ut ofEcia, quae liberis later legislation, seem to cast the situation in a darker light. The colonus is
congruunt, merito servilis condemnationis conpellantur inplere. saidto be iuris alieni; if he attempts flight, he is to be chainedlike a slave.
We are reminded of some notable features ofthe later colonate - the land-
Regardless of who [the person] is with whom a tenant-farmer per- lord's superiority over his tenant and the colonus's kinship to a slave. The
taining to another [landlord] is found, he is not only to restore [the question to be decided is whether these resemblances are inherent in the law
fugitive] to his place oforigin but also to answer for the tax liability of3 32 or becameapparentonly whenoneis familiarwithlater lawson the
[that the fugitive incurred in] the time [of his absence]. It is also colonate. On closer scrutiny, these passages deserve a far less ominous
fitting that those tenants who attempt flight be bound in irons as readingthanthey have generallyreceived.
though they were slaves, so that by virtue ofa slave's sentence they Ifiuris alieni in our text had its normal meaning in the law ofpersons,
be forcedto fulfil the dutiesappropriateto freemen. 12 distinguishingpersons sui iuris from those in the potestas ofothers, then we
would have to infer that a free man (regardless ofhisfamily status) fell under
Thislawhasbeensouniversallyinterpretedto documenttheexistence
ofthe boundcolonateunder Constantinethata long argumentis neededto 14 The inscriptions from Asiana do not indicate any sort of systematic enrolment of
demonstrate that it does nothing of the sort. 13 The colonus mentioned here tenantsonestates(seeAppendixl); Ulpian'smodel declarationfor estatesmakestheinser-
is specifically said to be "fi-ee, " to have the "duties" ofthis status, and to be tion of tenants optional and confines it m any case to those tenants responding to the
authorities in lieu of the owner (above, ch. iv, n. 2); CTh 11. 1. 14 (371$) distinguishes
saddled with "tax liability. " The law does not abrogate his right to change tenants registered on estates from those registered in their own right, and CTh 11. 1.26
landlords in the normal, legal ways, nor does it affect the landlord's right to (399) indicatesthat registry on estatesis still not required.Regardingorigo^Ulpian speci-
evicthim. Theorigothatthetenant cannotfleeishispatriaor placeofregistry, fies (Dig. 50. 1. 30) "Qui ex vico ortus est, earn patriam intellegitur habere, cui rei pub-
licae vicus ille respondet." The term origmar'ws, whichwould later designate a type of
colonus, refers in CTh 4. 12. 3 (320) to a (slave?) cultivator of fundi patrimomales (i. e.,
previousassessmentwasthe standardoffairness(above, ch. iv, n. 15), it appearsto form imperial land). The first association of origo with the penates of a cohnus occurs in 365,
the basisfor the medieval ideaofthe "nouvelle (et mauvaise) cofitume." CJ 11. 48. 6. On origo, seefurther CTh 12. 1. 12 (325), about curials; Gaudemet, "Constan-
12 CTh 5. 17. 1. Clyde Pharr, The Theodosian Code (Princeton, 1952), translated apud as tin et les curies, " pp. 57-8, illustrating the relationship oforigo to public burdens. Signifi-
in the possessionof"; origo as "birth status"; would-befugitives "must be boundwith candy,the Interpretatioto CTA 12.1.2 substitutes condicionascendifororigo (cf. the similar
chains and reduced to a servile condition, so that by virtue of their condemnation to alteration below, n. 25); this reflects what the law had become in the course of the
slavery" etc. fifth century.
13 B.g., Clausing, Cobnate, p. 17, "The first certain reference in the Codes to coloni Since neither a change of landlord nor eviction implied that a tenant must aban-
bound to the soil ... "; Marquardt, Staatsverwaltung, 11, 232; Jones, "Colonate, " p. I; don hisorigo, the law of 332 cannot be interpreted to sayanythingon the subject.
Roger Remondon, La crise lie I'Empire romain, Nouvelle Clio 11 (Paris, 1964), 147-8; 15 Since the opposite is often implied, it is worth pointing out that cohnus, like many
A. Piganiol, m. Journaldes Savants (1955), p. 12, "les textes ne connaissent [Iecolonat] que other categories in the laws, isby no means a hermetic designation. Legallanguage cannot
lorsqu'ilestdcji tresrepandu." Inorderto sustainthecontentionthat theboundcolonate bedirectlytranslatedintothat ofsociology. A tenantowninglandin hisownrightwas a
proceeded from imperial land, Emilienne Demougeot, De['unite a la division deI'Empire colonus in laws about tenants, a possessor or plebeius in other laws, and other things else-
romain (Paris, 1951), p. 46 with n. 93, mistakenlyreferredthe law of332 to colon!ofthe where. Even at a later date, one cannot reconstruct a colonus s life on the sole basis of the
resprivata. Other than this, I haveencounteredno one questioningthe interpretation of laws about the colonate; they tell us almost nothing, for example, of his relations to his
thislaw.
peers.
 72 CapU t AND COLONATB
FROM CapUt TO COtONATB 73

the potestas ofhis landlord by virtue ofundertaking to cultivate a tenancy; legislator had in mind when he spoke of servilis condemnatio. The punishment
becoming a tenant would have beena form ofcapifis deminutw. ^ Thisisnot of slaves is to resemble that of humiliores:
thecasein 332.Thetendency ofmodern commentators towards anachronism
hasobscured the small but vital point that alienus in this context cannot con- For the offences for which a free man is cudgelled, a slave is ordered to
trast with an implicit suus, meaning the colonus; it must contrast with quis- bewhippedandis [then]restoredtohismaster;andforthose[offences]
cumque, the other landlord. For a colonus, by definition, always works for for which a free man, after cudgelling, ishanded over to forced public
someone; colonus sui iuris, in the sense that ius is used here, would have the labour (opus publicum), a slave is ordered to berestored whipped to his
absurd meaning of "someone who ishisown tenant. " Besides, therest of the master under sentenceofchainsfor the time period ofhis [sentence].
law points consistently in the direction of the tenant's independence. The If he is ordered to be restored to his master under sentence of chains
fault ofour colonus isnot to flee from the landlord but to abandon hisongo and the master will not take him back, then [the slave] is ordered to
andhisobligation to thetaxcollector; again,whenwearetold ofthetenant's be sold, and if he finds no purchaser, he is ordered to forced public
attempting flight, the mischief envisaged is iiot towards his landlord but labourin perpetuity. 20
towards the state. As used in the law of 332, the words iuris alieni serve
simplyto specifytherelationship ofthecolonustotwolandlords,theone(A) By implication, the law of 332 allows the poena vinculorum to be applied to
fromwhomhehasfledandtheother (a) forwhomheiscurrently working: an apprehended fugitive colonus, but it is completely silent on three vital
owing to his prior contract with A, the tenant is iuris alieni in respect to details: who is authorized to apply the penalty, whetherjudicial proceedings
landlord B. Onewould say, intranslation, not that thecolonus is "dependent" or summary action are envisaged, andhowlong the sentence isto last. Owing
onanotherlandlordor "belongs"to him, but thatheis "undercontract"or to absenteeism on the part of landlords, as well as the interests in tax pay-
'obligated" to someone else, or, even more indeterminately, that he ments ofthe municipality and the tenant's fellow villagers, the permission to
pertains to another. The legislator's intention is simply to describe the "bind in chains cannot be assumed to have been granted exclusively to
factual situation of a fugitive tenant, and not to make the first landowner landlords. 21The penalty is harsh enough to have called for supervision in its
appear to be his tenant's overlord. " Another sixty years would pass before enforcement, but by whom? Besides, even a severe state rarely punishes
the laws stated that a colonus was in the potestas ofhis landlord and was not attempted flight with a life term on a chain gang. Though possible, such a
allowedto haveanythingsuiiuris.ls readingofthe lawis too grossto beadoptedin theabsenceofconfirmation
Although thesecondpartofthelawunquestionably tendstoassimilate in laterlaws.Theobjectofthelawisto retaintaxpayersforthecommunity,
thelot ofcolonitothatofslaves,it cannotbyitselfdocumentthebindingof not to increase the capital oflandlords by turning their tenants into slaves. As
tenant-farmers to the soil. 19 A passage from the Digest clarifies what the a result, the terms of the law cannot be taken as self-evident. A moderate
reading would bethat a sort ofcitizen's arrest offugitives isbeing authorized,
16 R.W. League, Roman Private Law,zndcd. (London, 1951), pp. 77, 102; Buckland, complete with chains, until such time as they give acceptablesecurity that
Text-book, p. 134.The classicassertionofthe independenceoftenants occurs in Gaius
Inst. 4. 153, "Possidere autem videmur ... si nostro nomine aliquis in possessione sit, licet they will indeed fulfil the duties offree men by remaining in their origo.22
isnostro iurisubiectus nonsit,qualisestcolonus etinquilinus"(FIRAS11, 184).Thisruleis
directlyexemplifiedin CTh4.22. 1 (326),wherecoloniarelistedalongsidepropinquivel were simply enslaved. The words as such might be construed in this way, but Pharr's
amid. Obviously, iusinthe lawof332 isnot usedwiththesameprecision asinGaius, but interpretation is all but impossible to substantiate in the other laws about the colonate.
thatishardlya singularphenomenon;cf.,forexample.Dig.332.28(Paulus),wherewe 20 Dig. 48. 19. 10. pr. (Macer). The passage begins: In servorum persona ita observatur,
would also translate ius as "obligation." ut exemplo humiliomm puniantur."
17 One may speculate that the legislator deliberately uses iuris aUeni in lieu of colonus 21 On the interest of convicani m the tax payments of the other tenant-farmers of the
alienus in order to avoid the echo that the latter expression would have to servus alienus, village, CTh 11. 24. 1 (360).
a very common compound (Dig. 28. 5. 3. 1 mdpassim). 22 The sort of security envisaged here is of the kind documented in late fifth- and
l8 CJlI.52. 1. 1 (393); ii. 50.2 (396). sixth-century papyri: E. R. Hardy, The Large Estates of Byzantine Egypt (N. Y., 1931),
19 UnlessoneadoptsthereadingofPharr(above,n.12), accordingto whichfugitives PP.75-7;JohnsonandWest, ByzantineEgypt,pp. 29-31; A. Segre, "Colonate," p. Ill;
 74 CapU t AND COLONATE FROM CapU t TO COtONATB 75

Anything more severe seems excluded by subsequent evidence. The forcible but also the local authorities and his fellow-villagers. What we have in 332
restraint of fugitives is vaguely echoed in 371, in a statement that runaways is a taxpayingtenant; the boundcolonate is still over thehorizon.
'arerecalled and subjected to penalties andchains. "23The allusion isa literary In closing this long commentary, it is important to point out that
echo ofthe terms of332 rather than a precise legal statement, and later laws legalevidencedocumentsonlylegalfacts.A discussionoftheboundcolonate
offer no clarification. They explicitly say that the condition of coloni is on the basis of legal evidence cannot pretend to present the whole picture.
similar to that ofslaves, but never again mention binding with chains, much Yet, whether bound or not, tenancy itselfis a legal condition, not a socio-
less enslavement. 24 The same reticence is manifested by the fifth-century logicaloragrarianclassification.Nodoubthistorianswishtoknowmorethan
Interpretatio to CTh 5. 17. 1, which avoids all reference to punishment; it legal rules, and they are wary of too closely correlating law and human
simply says: "[The colonus] who refuses to be what he was born [to be] is realities. Some wish to argue that the bound colonate had a long past, that it
restored to service. "25 The rhetorical vehemence of 332 never turned into constituted a gradual development, and so forth; and a little evidence is
enforceable law. available to support such views. 28 Nevertheless, a strict construction of the
In sum, the tenant-farmer was "bound, " but only by public obliga- laws is desirable and should not, from the first, be modified by sociological
tions, much as the contemporary curial was already denied escape to "the or agrarian considerations inferred from other sources. Precision is worth
bosomofverypowerfulhouseholds," tothearmy,to civilservice,andtothe attaining in the few categories of evidence where it can be obtained. The
Senate. 26The tenant was a free man, with the corresponding burden oftax result will admittedly be one-sidedly legal, but it has yet to be proved that
liability; in order to answerto the tax collector, hemust remain in hisorigo the bound colonate ever corresponded to a human situation rather than to a
and was forbidden to depart beyond thejurisdiction he lived in (so indeed legal one, or that it entailed harsher constraints than contractual tenancy.
was the landowner forbidden to transfer his professio from one city to The boundary between "free and "bound tenants is not one between
another);27 ifhe attempted to depart, he wasto be forcibly restrained - by happiness and misery for peasants or between good and bad economy. It
whom isnotspecified- untilhechangedhismind to theextent ofaccepting marks only a step - an important one - in a process oflegal definition.
his lot. Moreover, the receiver of a fugitive tenant-registered-for-taxation By 361 the situation of tenant-farmers had somewhat evolved. 29
committed an illegal act, by which he incurred the obligation ofpaying, or Senators had presumably been held responsible for the taxes ofany oftheir
perhaps only guaranteeing, the fugitive's unpaidtaxes, in additionto having tenants who had fled. As a measure of relief - part of an extensive statute
to return him. Those entitled to recover the colonus and back taxes from the
person illegally giving him shelter included not only his original landlord
28 For a typical exposition, see Ernst Stein, Histoire du Bas-Empire, tr. J.R. Palanque, I
(n. p., 1959), 28-30. Patchy and scattered like all the evidence for institutions in this epoch,
and Lombard charters: Schiaparelli, Coil. dipl. Long., nos. 55, 57, 8$, 104, 139, etc. Cf; that for the "gradual"development ofa dependent peasantryforms a sort ofobligatory
Goflfart, "From Rom. Taxat. to Med. Seign., " p. 182 n. 86. My suggestion is not that canon, often repeated but rarely if ever reassessed in its constituent fragments. Cf. the
such charters were already drafted in the early fourth century, but only that guarantees preferable position of Marc Bloch, who set imperial laws at the forefront of his exposition
of residence by colon!, in whatever form, were exacted at that time. but thought it necessary to add. The system of the colonate is only intelligible if we sup-
23 CJ ll. 53. 1, revocati vinculis poenisque subdantur." pose that there existed before it a sort of embryo seignemie" (Comb. Econ. Hist., i2, 2515-
24 CJ ll.52. 1 (393), 11.50.2 (396), "quadamservitutedediti"; 11.48.21 (530). 60). Even this statement may betoo directly oriented towards the known future portrayed
25 Ipse vero, qui noluit esse quod natus est, in servitium redigatur. " The servitium by the ninth-century polyptychs.
mentioned here does not mean slavery any more than the quaedamservitus ofCJ 11. 50.2 29 The two laws mentioning cotoni in the intervening years do not bear on the question
does. Note the interesting term "servitium publicum" in Gregor. Turon. Hist. 3. 15, whether or not colon! were "bound. " On CTh 13. 10. 3 (357), see below, p. 78. CTh
which I comment on, "From Rom. Taxat. to Med. Seign., " p. 187 n. 106, and Md., 11. 24. 1 (360) is the first law against patrocinia in Bgypt: honorati up to the rank ofdux,
p. 185 n. 97. who have interposed themselves between cohni and the devout" performance of their
26 CTh 12. 1.6 (3l8s), 10-11 (325), 13 (326), 14 (i'W.), 18 (ibid. s). Cf.E.Beaudouin, duties to the state, are obliged to reimburse the convicani for what they have expended
Lesgrands domaines dans {'Empire romain (Paris, 1899), p. 84. while their consortes abstained. Since this obviously relates to taxes and munera, it confirms
27 CTh 11. 22. i (346). the status of colom as free taxpayers.
 T

76 CapU t AND COLONATB FROM CapU t TO COLONATE 77

confirming senatorial privileges - senators wereguaranteedagainstdisbursing landlordsfor the debts oftheir departedtenants, leaving it to them to trace
Jiscalia for fugitives, provided they did not possess any of the runaway the fugitive andproceed against the retentator for recovery ofwhat they had
tenants own lands.30As wasnoted a few pagesago, the law of361 antici- been made to pay. In short, though landlords were reluctant to develop
patesby teayearstherulingthatdividedtenantsintotwocategories,accord- paternalistic relations towards their tenants, they were pressed by local tax
ing to whetherthey formedpart ofa bndlord'sprofessioorhaddeclaredland collectors into becoming willy-nilly the guarantors of tenant debts to the
oftheir own, howeversmall.31Thetendencytoequatetaxliabilitywithland state, except in caseswhere the tenant in questionhad declared land ofhis
was already clearly marked. own to stand as pledge for taxation. The complex system that assigned
The law of361 illustrates the pressures leading towards the binding of liability to individual citizens was on the way to being superseded by the
tenants as well asthe attitude oflandlordstowards the process. As the con- simplificationthatassignedliabilityto land,turningthecaputfroma personal
textimplies, senatoriallandownershardlywelcomedanincreaseinresponsi- charge,incumbent interaliasupontenant-farmers, intoa realchargeweighing
bility for the tenants oftheir estates. On the contrary, they sought asa privi- uponthe estatethey cultivated.
lege the right to be able to turn away liability for the taxes offugitives. It is In the same year that, in the East, Valensdistinguishedlandowning
worth keeping in mind that late Roman landlords showed no particular colonl, who answered to the usual collector, from coloni whose taxes were
desireto drawcloserto their tenants, asfathersto a family. Wealthallowed exactedbythelandlordor hisagent,a comprehensivestatutewasissuedwith
absenteeism; there was no point in having tenants if they had to be closely regard to the tenant-farmers of Illyricum.34The law is in three parts: first,
supervised. 32The authorities, however, looked instinctively to the landlord the amendment ofa law of 357 that had provided that when tenanted land
as the man responsible for his tenant; the chief collector of the tenant's was sold the tenants must be conveyed with the soil; second, a revised and
private debts was expected to assist the state in collecting his public debt. 33 developed version ofthe law of332; and finally, a set ofnewrules, super-
Presumably the law of332hadafforded a measure ofprotection to landlords seding those of 332 for present and future, that serves to document the birth
against these unwelcome practices by explicitly shifting responsibility from of the late Roman colonate. 35
their shoulders to those of the person receiving the fugitive. But this
was only limited relief. Authorities might continue to proceed against
34 Two laws relating to cofoni were issued in 365. CTh 5. 19.1 reinforces CJ 7. 32.5
(above, n. 5 *)by forbidding tenants to sell any property of theirs without their landlord s
30 CTh 11. 1.7 (ad senatum): "Conpertum est pro colonis profugisad exsolvendavos consent. Though often represented as an oppressive measure illustrating the tenant's
fiscalia conveniri. lubemus igitur, si nihil ex eorundem terris senatorum quemquam semi-servile status, this law merely extendsC] 4. 65. 5 (223), whichassertedthat when a
possidere constiterit, ut nulla cuiquam pensitandi pro his qui aufugerint necessitas farm or house was leased, what the lessee brought with him was obligated to the land-
inponatur. For the other parts ofthis law, see Seeck, Regesten, p. 208. lord pignoris iure (cf. Dig. 43. 33. 1). The requirement of consent was not a disability
3i Above, p. 64. to the tenant (in whoseprosperity the landlord hadan obviousinterest) but a safeguard
32 According to the agronomists, free tenancy was the manner of managing one's against the tenant's selling the landlord's property: CJ 7. 32. 5 (290-2); cf. 7. 30. 1 (226),
outlying possessions; and, owing to fragmentation, most private estates consisted of 39.2 (365)1 protecting landlords againstusurpation oftheir land by virtue ofpraescriptio
scattered pieces - ideal for lease and tenancy: K.D. White, Roman Farming (Ithaca, N.Y., tempons.
1970), PP. 404-5. On the prevalence of absenteeism, Jones, LRE, pp. 772-3. After all CJ 11. 48. 6 specifies that, in dealing with fugitive coloni and inqi iilini (adscripticios
what was the point of being rich unless it allowed extended residence in cities and is an interpolation), provincial governors are not to admit any distinction ofsex, munus,
capitals? Institutions like the government and churches were equally "absentee" as and mndicio; they must compel them to return "ad antiques penates, ubi censiti atque
landlords- a point that is too often forgotten. * educatinatiquesunt. " The law is obviouslyrelatedto that of332, ofwhichnothing had
33 The privilege that gave the state priority over other creditors (above, ch. in, n. 9) been heard in the interval. It suggests laxity in enforcement on the part of governors, but
wouldtendto conflictwiththelawsabsolvinglandlordsofresponsibilityforthepersonal also extendsthe interpretationofthe tenants' origo. Are "registeredandraisedandborn"
debtsoftenants: 074. 10. 11 (294), 11.48. 8.2 (371). Cf. Dig.49. 14.46.9, whichprovides regarded asa package or asalternatives to one another? This law iscertainly anindication
that, in caseswhen "multi fisco fraudem fecerint, " the procedure isthat "pro non idoneis that, after a generation, the government was renewing its interest in coloni. It may
qui sunt idonei conveniuntur. " One also wonders to what extend peasant liability to the represent a highly significantmoment in the history ofthe colonate (below, n. 51), but
state impeded governors from enforcing lease payments, as they were instructed to do, could not be seen in this light if it were not for the laws of 371.
C] 4.65. 17 (290); aslong ascolon!weretaxpayers,therewasa notableconflictofinterest. 35 CJ 11. 48. 7, h.t. 8, 53. 1. A fourth extract from the law, CTh 9. 3. 5, intensifies the res-
 78 CapU t AND COLONATB
FROM COpU t TO COLONATB 79

A commentary on thisimportant statute should beginwith the section with tax liability, the government sought to guarantee him the labour force
requiring that landretain its cultivators when sold. The original law on this he needed to keep the fields productive.
subject (357) hadprovided thattheseller ofpraedia couldnot privately agree The law of371, renewingthat of357, wasevenmore ofa limitation
with the buyer that he, the seller, should keep the tenants and transfer them on landlords than its predecessor had been. The object now wasto amplify
to other places" of his; "for if [the sellers] believe that the tenants are the prohibition of 357 affecting coloni by extending it to embrace rustici
advantageous, they should either retain them along with the fields or leave censitiqueservi, "agriculturalandregisteredslaves." Thesetoo wereto beleft
themtobeofadvantagetoothers,iftheydespairofthesefields'beingprofit- in place when an estate was sold, and if they were moved, they and their
ableto themselves. "36Thelandowner wasdeniedanyflexibility inmanage- progeny could be reclaimed as of right by the buyer and his heirs. 39 The
ment; he wasentitled only to retainhisland andits tenants or to sell them. prerogative ofowners to treat their slaves as chattels to be moved or sold as
Rather thanaffecting tenants, the measure restricted the freedom ofsales,as they wished was sharply curtailed; the settled and registered slave now
hadalready been done by the famous law ofland sales of3 3 7. In classical law, belonged to the field rather than to his master. 40 Four decades earlier, the
whathappenedto a tenancywhenownership ofthesoilchangedhandswas mancipiaadscriptacensibushadbeenconfinedtotheprovinceoftheirregistry. 41
a matter ofsome doubt andcomplexity; "... the effective plan wasto make It wasa long step from this to the measure of371, fixing them immovably
the colonus a party to the arrangement. " The landlord could either transfer to their patch ofcultivation - a striking illustration ofthe social effects of a
thetenants from a farm hesold to another heretained, or convey thetenants' 'land tax, " when assessed on land itselfrather than on persons.
leasesalong with the land; but the consent ofthe coloni wasmandatory if In its two sets ofprovisionsaffectingthe status oftenant-farmers, the
either form wasto bebinding upon them. 37What happened in 357wasthat statute of 371 made an implicit distinction between the past, governed by
thegovernment madeoneofthesecourses ofaction obligatory bydecreeing the rules of 332, and the present and future, to be governed by new rules.
thattenantshadto betransferred withtheland, whethertheyconsented or Sincethe law of332establishedthat colonicould berecovered from the land-
not. The fiscal motive ofthe law is plain; in much the same way, twenty lords to whom they had fled, it seemed necessary to augment it with a set of
yearsbefore, it hadbeenorderedthattaxliability must invariably betrans-
ferred in any land sale. 38After ensuring that the buyer ofland was saddled
39 CJ 11. 48. 7: Quemadmodum originarios absqueterra, ita rusticos censitosque servos
vendi omnifariam non licet. Neque vero commento fraudis id usurpet legis illusio, quod
m ongmarus saepeactitatum est, ut parva portione terrae emptori tradita omnis integri
ponsibilityofcommentariensesm guardingprisoners,"namipsum[scil.commentariensem] fundiculturaadimatur.Sedcumsoliditasfundorumvel certaportioadunumquemque
volumuseiuspoenaconsumi,cuiobnoxiusdocebiturfuissequifugerit." perveniat, tanti quoque servi et originarii transeant, quanti apud superiores dominos et
36 CTh 13. 10.3 (357): "Si quis praedium vendere voluerit vel donare, retinere sibi
transferendosad alialoca colonos privatapactionenon possit. Quienim colonos utiles possessoresvelinsoliditatevelinpartemanserunt:etemptorpretiumquoddederitamis-
sum existimet, nihilo minus venditori ad repetendos serves cum agnatione eorum vindi-
credunt,autcumpraediiseosteneredebentautprofuturosaliisderelinquere,siips;sibi catione concessa. Et si aliqua denique ex causa dissimulaverit legis usurpare beneficium
praediaprodessedesperant." It is worth notingthat the measurelaysdownnew law. atque iste sub hac taciturnitate decesserit, et heredibus eius et contra heredes emptoris
Personswhohadformerlyenteredintosuchagreementsarenottreatedasthoughthey vindicationem damus longi temporis praescriptione submota: mala fide namque posses-
had committed anything illegal. The guiding precedent appears to have been CTh sorem essenullus ambiget, qui aliquid contra legum interdicta mercatur. " The sharpness
II.I.4 (337),whichprovidesthatwhensomeonebuysfroma lesseeofimperiallandthat ofthe terminology in comparisonwith that ofthelawof357is particularlystriking.
privatelandofhisbywhichtheleasedlandissustained,thebuyerbecomesobligatedfor Owingto theearlier usageoforiginarius (above, n. 14),it isnotcompletely clearwhether
theburdensoftheleasedimperialland.Landthat"sustains"otherland,thoughpossibly thosereferredtoherearea categoryofpersonsorofslaves;I inclineto regardthewordas
nomorethananabstracttaxconcept, mayreferobliquely toa villageorothergrouping a reference to the coloniofthelawof357,whohavenow acquiredthisname byvirtue of
ofcultivators' houses; seeCJ11.48. 7. 1 (below, n.39), where therelation oflandto culti- the prominence oforigo in the law of332.
ya tors is spelled out. As a result, the law of 337 effectively forbade contractors of imperial
40 The impact ofCJ 11.48.7 wasso great that it wasinterpretedto forbidlandowners
landfrom sellingcultivators apartfrom thefieldstheyworked. Thisbecamegenerallaw from ever moving cultivators from one part to another of their estates. They were
twenty years later.
specifically guaranteed this right, CJ 11. 48. 13. 1 (400); renewed with some vehemence,
37 Digest 19.2. 32, with the discussion ofBuckland, Text-book, p. 502. NValent 35. 1. 18 (452). The stability guaranteed to coloni (add CTh 5.6.3 [409]) was
38 Above,ch.iv, n. 61.It was,infact,therequirementmadethenthatlandsalesshould eventually understood to extend to agricultural slaves, but this was denied: Edictum
bepublicthatallowedthegovernmentto imposetheadditionalrestrictionsregarding
cultivators.
Theodorici 142.
41 CTh ix.3.2 (327).
 80 CapU t AND COLONATE
FROM COpU t TO COLONATE 8l

criteria defining whether the receiver of the fugitive had knowingly or it alone suffices to determine that a crime hastaken place that hekept
unwittingly violated the law. Hisguilt or innocence wasto bedetermined by for himselfa person unknown to him. 43
theterms oftenancythathehadimposeduponthecoknus.Ifthetenanthad
been set to cultivating solely for the landlord's profit, or had been given a All the terms ofthe old equation have been changed. To begin with,
piece ofland to develop with no stipulation about the price ofhis labour, there is no more question ofregistry or capitatio. The stability ofa colonus
then the landlord was to be regarded as guilty, and subject to having the derives from where he was born andhis paternal family lived; the "fields"
unpaidtaxesofthe fugitive exactedfrom him. If, however, normal terms of (rus) that hecannot abandon arevirtually equivalent to histenancy. Gone is
tenancy hadbeengranted - namely, a sharecropping agreement or wagesfor thecompatibility withcontractual tenant farmingthatwasstillpresentinthe
work done - then the landlord wasinnocent, and the tenant wasresponsible laws of332 and 357. The second clause supplies the reason for the change:
for his own arrears. 42The general context ofthis section, which applied only thecolonusno longer paystaxesto thestate.Laterlawswould speakinthis
to the past, continued to be that ofthe law of3 3 2: tenant-farmers personally context of the abolition of capitatio humana, or of disregarding the "tax
responsible as free men for taxes to the state.
liability ofa soul. "44The sameisimplicitly affirmed here.The original caput
The new regulations of 371, applicable to the present and future, to theextent that it embraced real property hasbecome identified with therus
completely revise the status oftenant-farmers:
wherethetenantresides, whosetaxliability iswholly incumbent upon the
landlord. The rest ofthe old caput, namely the earning capacity that had
Wejudge that tenant-farmers and cottagers in Illyricum and qualified the tenant as a contributor to the annona, is simply abolished. 46
neighbouring regions cannot havepermission to depart from thefields But not without replacement: the personal obligation of the colonus to the
in which it is certain that they dwell by virtue ofbirth and paternity. annonais turned into an obligation to the fields, that ofkeeping them culti-
Let them do service to the lands not by virtue of indebtedness for vated and productive. The legislator speaks ofthis duty asarising fi-om the
taxes but by the obligation and intitulation ofcultivators, such that, women et titulum colonorum. Later, it would occasionally be called the ius
if they depart or pass over to another [field], they are recalled and agrorum, the right ofthe fields to be fi-uctified by their cultivators. 46
subjected to penalties and chains. The receiver of a colonus who has ceased being a taxpayer can no
Those who suppose that an alien and unknown person is to be
received may expect a penalty both in compensation ofworking days
and in damage that was done to the places that [the tenants] deserted, 43 CJ ll. 53. 1: Colonos inquilmosqueper Illyricum vicinasqueregionesabeundirure,
m quo eos ongmis agnationisque merito certum est immorari, licentiam habere non
and a fine whose amount we leave to the determination ofthe judge. possecensemus.Inseryiantterris nontributarionexu, sednomineet titulo colonorum. ita
Consequently, even the master ofan estate in whichan alienis shown ut,siabscesserintadaliumvetransierint,revocativinculispoenisquesubdantur,maneatque
to have been is forced to undergo punishment to the extent of the cos poena, qui alienum et incognitum recipiendum esse duxerint, tam in redhibitione
operarum et damni, quadlocis quaedeseruerant factum est, quam multae, cuiusmodum
transgression, and there is no occasion [for pleading] ignorance, for in auctoritateiudiciscollocamus:ita at etiamdominusfundi,in quo alienusfuissemon-
strabitur, pro qualitate peccaticoercitionem subire cogatur necsit ignorantiae locus, cum
adcriminisrationemsolumilludsufficiat,quodincognitumsibitenuit. " Therearefurther
42 C/ll.48. 8: Omnesprofugiin alieuolatebrascollocantescumemolumentistributariis, provisions about slaves and freedmen, the latter of whom fall under the same rules as
salva tamen moderatione, revocentur, scilicet ut si, apud quos homines reperiuntur, liberi coloni.
alienos esse noverant fugitivos et profugis in lucrum suum usi sunt, hoc est sive exco-
44 CJ 11. 52. 1 (393s); CTfe 11.20. 6 (430). 45 As qualified above, ch. iv, n. 71.
luerunt agros fructibus dominis profuturos sive aliqua ab isdem sibi iniuncta novaverunt 46 CTh 10.20. 10. 1 (380); 5. 18. 1.2 (419). Valentinian m speaks ofius etnomen colonorum
nec mercedem laboris debitam consecuti sunt, ab illis tributa quae publicis perierunt {NValent31.1. 1 and6). Augustinesuppliesperhapsthemostprecisedefinition(Decivit.
functionibus exigantur. Ceterum si occultato co profugi, quod alieni essevidentur, quasi Dei 10. 1.2), coloni, qui condicionem debent genital; solo propter agriculturam sub
suiarbitriiacliberiapudaliquemsecollocaverunt autexcolentesterraspartemfructuum dominio possessorum" (which echoes adsolumgenitale in the law of 419 above). Ch.
prosolodebitamdominispraestiteruntceterapropriopeculioreservantes,velquibuscum- Saumagne, Du role de Yorigo et du census dans la formation du colonat romain,"
que opens impensis mercedem placitam consecuti sunt, ab ipsis profugis quaecumque Byzantwn,12(1937),510-12,rightlystressedtheexpressioniusagrorumasindicatingthat
debentur exigantur: nam manifestum est privatum iam esse contractum. " The criterion
thetenant'sobligationslayelsewherethantothelandlord.Butthisiusdidnotgiverise
of intent was introduced some years later, CTh 5. 17. 3 (386-75). to doctrinal commentaries; it was seldom mentioned.
 82 CapU t AND COLONATB FROM CapUt TO COtONATE 83

longerbeheldaccountablefor unpaidtaxes.Thisis thechangeincorporated long enough, in 371 and after, over the nature ofwhat the retentator now had
into the final clauses of the law of 371. In lieu of taxes, there is the debt to to payto confirm thattenant farmers hadbeenconsciously dropped from the
the fields, namely working days and damages; and, in view ofthe criminal rolls ofthose paying taxes to the state.
nature ofthe offence, a fine. A glance ahead at subsequent legislation proves Though the laws of371 may not have instituted the bound colonate,
that these refinements had no future. In 3 86, an eastern law takes away the they are the first to document the characteristic features of the institution.
initiative fromjudges by fixing the fine at six ounces ofgold for a private It ishere that we first seetenants exchanging tax liability for an obligation to
colonusand onepound for a tenant ofimperialland;a fewyearslater (393), the soil; where, in other words, the state bestows upon "the fields" the
thetotal penalty consistsinreturning the fugitive withhisfamily andbelong- unsecured debtsin respect to their personsthattenant-farmers incurred long
ings {-peculium), and paying two pounds ofgold, with no details about work- ago by registering capita, "unsecured shares ofassessment. " Since the bound
ing days or damages and no distinction between private and imperial. 47The colonate had yet to be extended to Palestine and Thrace, we have reason to
final enactment on the subject (396), whichmentions tributarii and inquilini conclude that, in371,it hadonlyjust comeintobeing. 61Thepressures leading
but not coloni, extends to receivers (detentatores} ofordinary runaway tenants to its establishment are implicit in the laws of 332 and 361, as well as in the
the samepenaltiesasfor receivingfugitivefiscalslaves,namelythereturn of eventual legal novelty ofthe nomen colonorum and ius agrorum. Tenants were
an additionalbody along with the fugitive and a fine oftwelve pounds of an elusive category of taxpayers. The tendency of tax collectors to hold
silver. 48 The subject ofcompensation then disappears from the laws without landlords responsible for the taxes oftheir tenants may have prevailed over
ever having been authoritatively reformulated. What rendered new rules the resistance oflandlords to this responsibility. 52 Ifcircumstances militated
superfluouswasthattheobligationsofthecolonustohislandlordwerequickly against a system ofassessment ofpersons and in favour ofthe direct assess-
identifiedwith tributaor annuaefunctiones, "taxes"servingasa synonym for meat ofland, howmuch greater still werethedifficulties ofkeepingtrack of
"rent," since obviously the landlord could nowavailhimselfofpublic pro- small, marginal proprietors than ofthe greater landlords. 63A further element
cedure for recovery oftax debts in forcing the colonus to pay what he owed. 49 to beaccounted for isthecommutation oftaxes inkind into taxesinmoney;
Theshiftinvocabularyrejuvenatedoldlawina waythatthelegislatorof371, the evidence for this practice, which was always an integral feature of the
too close to the idea ofa tenant as a real taxpayer, could not yet anticipate. fourth-century system, suggests that it had reached such great proportions
If the tenant's debt to the landlord consisted in tributa, one could fall right by the 3 Sos that deliveries in kind had come to be regarded as the minor
backupon the provisionsofthe law of332, makingthereceiverresponsible
for unpaid capitatio. As restated in the version of the fifth-century Infer- 5i The alternative is the law of365, CJ 11. 48. 6 (above, n. 34), addressed to Gaul. Ifthis
pretatio, the rule reads as follows: "If anyone knowingly retains an alien isunderstood to broaden andextend the definition oforigo in thelawof332, no additional
legislation would have been needed in order to "bind colon!. " The lawsof371 would then
colonusin hishousehold,let himrestore [thecolonus}to thefirst landlordand represent the codificationofthe practices ofthe Gallic prefecturewith a view to their
be forced to disburse [the colonus's] tributafor as long as he stayed. "50New extension to that oflllyricum. But this is conjecture.
conditions could be covered by old law, provided the landlord wastacitly 52 Above, p. 76 with n. 33.
53 In the calm ofthe tax records, the Empire appeared asanaggregate ofshares of assess-
castasa tax collector. Howeverthat may be, the government hesitatedjust ment, wgavel capita,thatwere all expectedto yield revenue. Reallandandreal people
were more intractable; land became unprofitable and was taken out ofcultivation, while
47 CTh 5. 17.2; CJ 11.52. 1.2. men died or moved away. Normally, losses were compensated by gains, but who was to
48 CJ 11.48. 12 (396s). The law stating the penalty on receivers offiscalslaves, CJ6. 1.7 conduct the delicate operation ofadjusting assessment to changing realities? Local author-
(371)- According to CJ 6. 1.4 (317), the receiver of a fugitive slave restored twofold or ities, who had formerly done this and who alone operated at the level where it could be
paidtwenty solidi.In CJ 11.48. 13 (400), coloniaresaidto beidenticalto inquilini,but only done, were not to be trusted; they might not befair (above, n. 9). Better to remove the
"quantum ad originem pertinet vindicandum." adjustmentfrom a politicalprocessanddependinsteadonimperiallawsthathemmedin
49 Annuaefunctiones, C] 11. 50.2 (396); capitationis sarcina, CJ 1.3. 16 (409); tributa vel taxpayerswithrestrictions ofsucha kindasto guaranteetheir yieldingrevenue to the
functiones, NTheod 7. 4. 2, 5 (441)- state. In this way, legislation would make reality more closely approximatethe ideal;
50 [nt CTh 5. 17. 1, Siquis alienum colonum sciensin domo suaretinuerit, ipsum prius real land and men would act like the assessment shares ofthe taxrecords. This isthe reason-
domino restituat et tributa eius, quamdiu apud cum fuerit, cogatur exsolvere. "* ing suggested by the pattern offourth-century tax legislation.
 84 CapU t AND COLONATB FROM CapU t TO COLONATE 85

component of the tax bill. 54 Tenants had qualified as contributors to the standing law, theintroduction ofthemeasureisjustified byanappealto im-
annonawhen it had been collected in kind; the economic situation did not so memorial traditionandimperial uniformity:
farimproveinthefourthcenturyasto enablethemto continuecontributing
when it was commuted to gold. The evolution oftax assessment, the evanes- Since, in the other provinces that are subject to the empire of
cence ofmen oflittle property, the tendency towards commutation: all these Our Serenity, a law laid down by the ancients detains coloni by a
urged that tenant-farmers be struck from the rolls oftaxpayers. Wisely or certain eternal right, such that they are not allowed to depart from
not, an anxious government simultaneously thought it necessary to bind those places by whose harvest they are refreshed or to desert those
them to the fields.
that they once undertook to cultivate; and since this does not bring
The law of371 wasnot anisolated enactment. Three additional pieces assistance to the landowners of Palestine; we decree that, in the
oflegislation allow usto trace further steps in the binding oftenants to their Palestines aswell, no tenant-farmer whatever may rejoice in indepen-
fields. A law for Palestine (393) shows us how matters were handled in a dence, asthough unfixed andfree, but onthemodeloftheotherpro-
region wherecapitahad not beenregistered; one for Thrace (396) caststhe vinces he is so held to themaster oftheestate that he cannot depart
tenants' confinementm terms ofa cancellation oftax liability; and one for without [bringing] punishment upon [his] receiver. The following is
Gaul (399) illustrates how, at length, the caput could be misconceivedas a added: that complete authority to recall him is granted to the master
'manner of retaining commoners" (ratio retinendae plebis). We need not of the land. 57
supposethatparticularlawswerenecessaryinall casesto changethestatusof
coloni. Providedtheir tax liabilitywasregisteredandprovidedregistry pro- A westerner like TheodosiusI may havebeenablewithout bad faith
ceduresweresohandledasto identifyorigowith estate, thefixity oftenants to think ofthecolonate asaninstitution ofgreatantiquity. The point hinged
could be brought about on the mere basisofenforcing the law of332; no upon appealing to the right ofrecalling tenants, which was more than fifty
further change was required than that tax liability should devolve from years old, rather than invoking the novel iusagrorum. w This is why the law
persons to land and that the courts should admit the principle of the ius for Palestine,moresothananyother, stressesassistanceto landlords.Modern
agrorum or colonatus. ^ The laws we have arose from particular circumstances commentators are often quick to conclude that, though thecolonate wasnot
that we cannot always identify. primarily instituted for the benefit of landlords, they are likely to have
The law for Palestine imposes colonary restrictions upon tenant-
farmersin anareawherecapitahadnever beenregistered,andthesubjection
ofcoloni could not beregardedashavingbeenprepared by their obligation Jones, LRE, p. 797 and above, ch. iv, n. 22 (the case of Palestine is similar). Deleage,
Capitation, pp. 148-9, inferredthe existenceofa headtax in Palestinefrom CTh 16.2.26
to the state. 56 In lieu of being treated as an extension or amendment of
(381), "quis enim eos capite censos patiatur esse devinctos ... ?" (referring to clergy).
As usual when censusis in question, the text speaks ofassessment, without specifying the
54 On this subject, seenow Cerati, Carac.annon., part I, whichsupersedesearlierwork, tax to be imposed on the censiti. *
such as the very limited discussion by Jones, LRE, pp. 207-8, 460-1. Mickwitz, Geld. 57 CJ 11. 51. i (3935): Cum per alias provincias, quae subiacent nostrae serenitatis
u.Wirtschaft,and Santo Mazzarino,Aspettisociali del 4. secolo (Rome, 1951),arguedcon- imperio, lex a maioribus constituta colonos quodam aeternitatis iure detineat, ita ut illis
tradictory theses on commutation (adaeratio) whose unsatisfactory character was well non liceat ex his locis quorum fructu relevantur abscedere nec ea deserere quae semel
stated by H.-I. Marrou in Gnomon, 25 (1953), 187-8. Cerati has usefully distinguished colenda susceperunt, neque id Palaestinae provinciae possessoribus suffragetur, sancimus,
adaeratiofrom transtatioin aurariampraestationem, a practiceheinferredfrom CTh11.20.6 ut etiam per Palaestinas nullus omnino colonorum suo iure velut vagus ac liber exsultet,
(43°).* sedexemplo aliarum provinciarum ita domino fundi teneatur, ut sinepoena suscipientis
55 It must be stressed, however, that the only evidence we have for such an evolution is non possit abscedere: addito co, ut possessionis domino revocandi eius plena tribuatur
CJ 11. 48. 6 (365), where origo seems to be identifiable with the peasant's penates. By con- auctoritas."
trast,thelawof361 (above,n. 30) setsa ceilingona landlord'sliability.Thereisnoreason 58 One might suppose that the less urbanized regions ofthe West witnessed the identifi-
to think that the evolution from 332 to the bound colonate was a smooth and direct cationoforigowithestate(asabove,n. 55) andthusallowedaninterpretationofthelaw
route.
of 332 such that the right of recall referred more directly to landlords than to com-
56 This appears to follow from CTh 7.6. 3 (377) and the present law (CJ ll. 5X. l); cf. mumties.
 86 CapU t AND COLONATB FROM CapU t TO COLONATE 87

welcomed it. 59 This may be questioned. Landlords might have gladly by right of birth, and though they appear fi-ee in status, let them
relinquished the doubtful advantage of immovable peasants if doing so be regarded as slaves ofthe land to which they were born, and not
would have won back the rights to move settled slaves, to raise rents, to evict have the occasion of withdrawing wherethey wishor of changing
unsatisfactory tenants, to attract the labourers oftheir neighbours, andto sell places. Their landlord may rightfully availhimself[intheir respect] of
landapart from tax liability - alltheprerogatives ofownership that, over the the prerogatives ofa patron andthe power ofa master.
years, they had been denied. 60 The colonate appears to owe far more to Ifanyone believes that an alien cultivator is to be received or
imperialpaternalismthanto a landlords' lobby. Nevertheless,thelawof393 retained, let him be forced to disburse two pounds gold to him whose
may be invoked to document the traditional view, if one chooses to do so. 81 fields the fugitive deprived of cultivation, and let him restore [the
Since the law of396 for Thrace presupposes that tenants were ab-eady fugitive] withall his personalbelongingsandprogeny. 62
fixedbutmightnowgeterroneous ideas,thebindingoftenantsmayalready
havebeen enforced by applying the old tax laws, asinterpreted in the light Inviewofa lawof386for Pontus,whichloweredcaputratesformen
of modified forms of registry and tax collection. What appears to have and women, it is conceivable that, in Thrace, a census humanae capitationis
occasioned the law of 396 was an abatement in taxation: on a conservative hadbeendrawnupwithinrecentdecades.68Butthisinferenceisnotmanda-
reading, secured tax shares were retained as the assessment base, while un- tory. In any region where liability had beenregistered in iugaandcapita,a.
secured shares were cancelled; alternatively, all capita were struck from the proportion ofthe capitarepresented real property, slaves, andlivestock, while
rolls, leaving iugaasthe sole basisfor taxation. the balance consisted of the contributive capacity of human beings. If
collectorshaddespairedofkeepingtrackofthesepersons,aswelltheymight,
Throughout the diocese ofThrace, the register of human tax they would long have sought the total assessment from the portion ofthe
liability is cancelled forever and only landed tax liability is paid. register representing luga and zugokephalai, "real" capita. The irrecoverable
And lest perchance thetenant-farmers, whoare [now] absolved residue, whose cancellation would be welcome relief, could acquire the
from the obligations of being taxpayers, should appear to have been noblenameofcensushumanaecapitationisfrom a legislatorpreferringliterary
allowed to wanderandto depart wherever they wish, let them beheld elegance to historical accuracy. 64These refinements mattered little to tenant-
farmers, whose status in fact now came under the aegis of more suitable
59 Jones, LRE, p. 796: It was the interests oflandlords, who were after all by and large legal rules than it formerly had. The confinement ofmen to the soil on the
identical with the official aristocracywho controlled governmental policy, whichpre- pretext oftheir being taxpayers was superseded by a confinement based on
vailed. Cf. Otto Seeck, "Die SchatzungsordnungDiocletians," Zeitschriftfur Sozial- their being "slaves ofthe land to whichthey were born, " the lus agrorum.
undWirtschaftsgeschichte, 4 (1896), 314, asserting that court favour was"der entscheidende
Factorder ganzenPolitik. " The ideathat the interests ofthe landownersofthe Empire
were homogeneous looks like a gross oversimplification; we seem to be dealing with 62 CJ 11. 52. 1 (396s), "Per universam dioecesim Thraciarum sublato in perpetuum
something as analytically precise as the modern polemical usage of "Big Business. " The humanae capitationis censu iugatio tantum terrena solvatur. Et neforte colonis tributariae
reasoning exemplified by Charles Wilson, "Government Policy and Private Interest in sortis nexibusabsolutisvagandiet quo libuerit recedendifacultas permissavideatur, ipsi
Modern English History, " in Economic History andthe Historian: Collected Essays (London, quidem originario iure teneantur, et licet condicione videantur ingenui, senri tamen
1969), PP. 140-55, offers an admirable corrective to such points ofview; seealso Wilson's terrae ipsius cui nati sunt aestimentur nec recedendi quo velint aut permutandi loca
apt remark, ibid., p. 93. habeant facultatem, sed possessor eorum iure utatur et patroni sollicitudine et domini
60 Slaves must be sold with the land, CJ 11. 48. 7 (371); no raises in rent, C] 11. 50.1 potestate. Si quis vero alienum colonum suscipiendum retinendumve crediderit, duas
(325); eviction ofcoloni forbidden on leased imperial land, CJ 11. 63. 3 (383); no seduction auri libras ei cogatur exsolvere, cuius agros transfugacultore vacuaverit, ita ut eundem
of tenants from origo, CTh 5. 17. 1 (332); sale of land apart from liability forbidden, cum omni peculio suo et agnationerestituat.
CTh 11. 3. i (3135) and Fragm. Vat. 35 (337?). Cf. Heitland, Agricola, pp. 394-5. As far as 63 CTh 13. 11. 2 (above, p. 60). Such a register of purely "human" capita would have
I know,we haveno directevidenceaboutthe preferencesoflandlords, a vast andvery been separate from the register ofsecured shares ofassessment [iiiga vet capita).
diverse category. 64 By a process ofattrition within a single register, the perishably human constituent of
61 As done byJones, "Colonate, " p. 6, and in Third International Conference of Economic the capita would have been segregated as bad or problematic debts from the capita vel
History, 1965 (The Hague, 1969), p. 99. itya secured by, or representing, real property.
 88 CapU t AND COLONATB
FROM CapU t TO COLONATE 89
However questionable their combination may be,censusandcapitatio itieswhenthelandchangedhandsor wasexaminedforsomeotherreason. 67
at least make an appearance in the law for Thrace. The final document to
be examined omits these terms altogether but may nevertheless illustrate the Presumably the petitioners hadproposed that one oftwo courses beadopted
in lieu ofspecial privileges: either capita should be registered on all estates
closing phasein the disintegration ofthe old caput. Thelawisaddressedto that had settled cultivalors, or assessment should be based on iuga alone; in
the Vicarius Galliarum:
other words, plebs should besettled on all estateswith corresponding liability
to the owner, or theadded liability, whichweighed only on some, should be
All privileges ofbeneficeshavingbeencancelled, YourHigh- abolished. On this interpretation, the complaint had arisen from the land-
ness is to order to be astrained to public obligations all the landowners
lords mistaken conception ofthe caput;they took it to be a ratio retinendae
oftheseprovincesfromwhichthecomplaint [nowbeingdecided]has plebis for which additional liability was incurred. The government, while
arisen or [to put it another way] in which this registry or manner of adopting the phrase, recalled the past with sufficient accuracy to turn down
retaining commoners is practised.
the appeal. It cancelled the privileges, which if at all justifiable had been
No onewill berelieved out offavour, nopersonwill bevexed granted on erroneous legal grounds, andreaffirmed existing practices. Capita
by the disadvantage ofinequitable assessment. Let everyone endure of assessment and registered tenants stayed where they were, even when
the same lot, provided, however, that if an estate is transferred to
registrydisagreedwithreality;andtheacquisitionofadditionalfixedtenants
another person - an estatein whicha definite number ofcommoners resulted in increased assessment. 68
hasbeenregistered - thenew owner is compelled to answer for the The origins of the bound colonate should not be confused with its
burdens ofthe object sold, for it appears that commoners are neither
future, which will not be explored here. 69Thesepageshavesought to show,
toberegisteredinallestatesnortobestruckfi-om[theassessmentof]
that [estate] to which, hereafter, they have oncebeenassigned. 65
6^ The laws called for fiscal inspection and revision ofassessment on various occasions:
The text is disconcertingly obscure, owing to our ignorance ofthe CTh11.3.2 (337),upontheacquisitionofregisteredslaves;fragm.Vat.35,whenproperty
wassold;revisionsofassessment,CTh13. 10.5 and7 (367, 371).Adjusters(peraequatores)
substance ofthe querimonia, which we must somehow reconstruct. Jones begin to be mentioned in 377 (CJ 11.62. 5), an indicator offurther centralizationofthe
inferredthatincertainprovincesofGaulthe"polltax"wasseparatedfrom tax machinery.
the"landtax, " whileinotherslandlordswereresponsibleforbothonbehalf 68 Another point to be recalled in the interpretation of CTh 11.1.26 is that, in the tax
terminology of Gaul, caputalone appears to have been used as the term for a share of
oftheirpkbs. s61suggestthathaecretinendaeplebisratioadsmptioqueshouldbe assessment, regardless ofwhether "secured" or not (see above, p. 49). What the law of
takenasanotherexampleofa late,circumlocutoryreferencetothecaput. 399 presupposes is that the sense ofcapt itis "animate" tax liability, regulated by a schedule,
Whatappearsto havehappenedisthis.Intheprovinces in question, hadlongfallen out ofusage.The ambiguity betweenthenowunique "landed"caputand
certain estates wereassessedin iugaonly; others in iugavelcapita;others still theobsoletebut residual"animate"caputwaswhatoccasionedthequerimonia.
69 The subject deservesre-examination,especiallywitha viewto the prolongationsof
had,byfavour, obtainedvariousmeasuresofrelief.Ifadditionalplebswere the colonate beyond the sixth century. See M. Pallasse, "Les 'Tablettes Albertini'
permanently settleduponestates, theassessment wasincreased bytheauthor- int&essent-elles Ie colonat romain du Bas-Empire?" Rev. hist. de droitfranf. etetrang., 4th
ser., 33 (1955). 267-81. There are important implications to the observation of H.J.
Scheltema,AndenWurzelndermittelalterlichenGesellschaft,11,DasostromischeReich(Oslo,
65 c ".l.2<5,"Omniamotoprivilegio beneficiorum possessoressublimitastuaprae- 1958), p. 87: "DasVerhaltnisder Potentioreszu ihren InquilinenundAdscriptitienist
cipiet UDiversos munenbus adstringi scilicet provinciarum,
earum ex quibus orta queri-
wesentlicheinVerhaltnisdesoffentlichenRechts;derParoikosistschliesslichebensowenig
maniaestautinquibushaecretinendaeplebisratioadscribtioqueservatur.Nullumgratia ein Sklave,alswir selbstdie Sklavender Steuerinspektorsind." It isworthassessingthe
relevet, nullum iniquaepartitioms vexet incommodum, sedpanomnes sorte teneantur:
impact upon colon! of the laws indicating the easy availability of land for cultivation
ita tamen, ut, si adalterius personam transferatur praedium, cui certus plebis numerus {CTh 11.7.4; CJ 11.62.5; 11.59. 11; CTh 13. 11. 13; 5. 16.34-5; 5. 12.3) and ofthe ones
fuerit adscribtus, venditi onera novellus possessor compellatur agnoscere, cum plebem allowingcurialsacquiringprivilegedstatusto install substitutes (CTh 12. 1.155; h.t. l6l,
constet non tam in omnibuspraediisadscribendamneque auferendamab co, cui semel
posthac deputata fuerit." 163, l6o, 181). One also wonders whether the colonate under Justinian really offers a
66 Capitatio, p. 92. sharpcontrastto theconditionsportrayedbytheFarmers'Law,asmaintainedbyP.Le-
merie, EsquissepourunehistoireagrairedeByzance, " Revuehistorique219 (1958), 63-5;
 90 CapU t AND COLONATB

on the basis of the laws, the social effects that taxation had upon the most VI
marginal layer of taxpayers. A study of the establishment of the colonate
complements whatevermay be learnedfrom studyingthetechnicaltermsof
assessment. When the laws are chronologically ordered, three parallel Conclusion
phenomena are observed at the close ofthe fourth century: the emergence of
the idea of the caput as "the assessment of a soul, " a gloss that is wholly
inappropriateto theearlierevidence;theintroductionofthe boundcolonate;
and the short-lived change in the penalty inflicted on the fugitive tenant's
retentator, from a refunding of taxes to a fixed fine. By their close relation-
ship, the three developments have the advantage of confirming the inter-
pretation of each one. The devolution of assessment to land alone entailed
the disintegration of the caput and the binding of the tenant to the soil as a The interest historians have shown in late Roman taxation stems partly from
substitutefor keepinghim ontheroll oftaxpayers.Wearetakenbackto the the prominence that the Codes give to this subject and its corollaries, but
extension oftaxationto a wider body ofcontributors under the Tetrarchy, even more from. the long-held suspicion that taxation was somehow related
aswell asforward to some ofthe effectsthat generaltaxationhaduponthe to the declineofthe Empire. A simplistic but traditionalproposition is that
internal life ofthe Empire. great states tax themselves into extinction, andthat Rome was one ofthem.1
More thoughtful consideration of the evidence about taxation suggests that
it is the most sensitive indicator we possess of the economic course of late
how can there be a contrast when one knows that the Justinian Code and Noveltae
remained in force? Antiquity.
The previous chapters are a preliminary step towards the understand-
ing oflateRoman taxation. Their object hasbeento sort out a terminological
puzzlethathaslong bedevilled scholars. Whattheyhaveshownbeyondthis -
the devolution of assessment from personal fortunes towards land itself -
is a phenomenon whose economic significance should not be hastily evalu-
ated. Beforerecapitulatingthesefindings,it isappropriateto stressthemore
general points that have been made on the nature oftaxation in the Roman
Empire.
Thefirst oftheseisthattheideasunderlyingtherelationsbetweenthe
taxpayer and the state in Antiquity difFeredprofoundly from the ones we
areaccustomedto. Whileour modeltaxpayerhasthepassivevirtueofobey-
ing the law, taking every legitimate advantage it affords him, the ancient
model was that of the civic benefactor, who gladly met more than his share
of the common annual burden and did not hesitate upon occasion to make

I SeeC. Cipolla, The Economic Decline ofEmpires (London, 1970), pp. 5-7, a sophisticated
version of this point of view. See also Charles Wilson, "Taxation and the Decline of
Empires, an Unfashionable Theme, " in Economic History and the Historian (London, 1969)
pp. 114-27, arguing that the new style ofeconomic history has gone too far in disregard-
ing publicfinance(with specificreferenceto theeighteenth-centuryDutcheconomy).
 92 CapU t AND COLONATB
CONCLUSION 93

inroads into hiscapital in order to benefit hiscommunity. There isno reason sive economy. 6 A corrective for this tendency involves taking account ofthe
to sentimentalize over this situation, or to exaggerate thealtruism oftherich. extension, normality, and long duration of these practices in Antiquity, an
Generosity towards the public purse compensated for private exploitation epoch whose economic context wasvery different from that ofearly modern
and thus constituted the basis for social concord. Munificent philanthropy Europe. "Liturgies" and munera appear to have been a form of taxation
was a component of civic virtue, and not additional to it. Moreover, the admirably adapted to the political order and economic circumstances of the
ethics of tax gathering, as then conceived, called for the rich to contribute times. The delicatequestionraisedby munerais not whetherthey weredesir-
firstandfor citizensofmediumandsmallfortunesto besparediftheneeded able but what level of government could benefit by mobilizing them.
total could be raised without them. The system endured as long as it did Municipalities could; in them, unremunerated services were a sizeable asset
because advantageous to all concerned. A historian of Roman taxes cannot on the balance sheet of public finance. But what use could the imperial
afford to forget this ideology of openhandedness, and one thing he must government make of such services other than those directly affecting its
explain is how it came to be effaced in the last centuries ofthe Empire.2 income? What effects were produced when the emperors encroached upon
Another distinctive trait of ancient taxation was the large place it munerain their efforts to widen the imperial share of public revenues? This
reserved for the contribution of unpaid public service: in the higher ranks problem cannot be overlooked in a history oflate Roman taxes.
of society, onewascalled upon to perform municipalofficesandfunctions; A third trait, embracing the first two, isthe municipal focus ofimperial
in the lower ranges, a variety of labour services. The latter were labelled taxation, which prevailed well into the fourth century. To envisage the
munera sordida as the Empire wore on; eventually, the authorities realized inhabitants of the early Empire as paying direct taxes on their lands and
that this adjective had done irreversible harm by encouraging wholesale heads to the emperors, asis often done, misrepresents the caseandmakes it all
evasion on grounds ofprivilege. 8 Yet public services ofa menial kind were as the more difficult to interpret the course of events after Diocletian. Rather,
traditional as those demanded of rich men. We find labour services in the the demands of the imperial government were addressed to collectivities,
charter ofUrso (44B.c. ) andin the African inscriptions about imperial saltus which were assured of sufficient sources of revenue to cover both internal
aswellasin lateRomanlegislation,whosenoveltyresidednotin theservices expenditures andthe amounts to beremitted to higher authority. Ina scheme
themselves asmuch asin the level ofgovernment prescribing them. 4 One of of this kind, local governments played the pivotal role; they were particu-
the economic reforms ofTurgot, in the closing years ofthe French ancien larly useful because their intimate practical knowledge of local wealth and
regime, was to replace the royal corvee with a money tax. 5 The instinctive circumstances made up for the deficiencies of censuses and bureaucratic
modern tendency is to reason that Antiquity should have acted in the same record-keeping.7 The "liturgical"spirit oftaxation,manifestedin voluntary
way;muneraofanykindareregardedascharacterizinga primitive orregres- generosity and unremunerated services, had a petite patrie as its favoured
terrain. By keeping such municipal patries at the centre of the tax process,
2 Fora briefoutlineoftheproblemofhowthespiritofpublicmunificencewasblotted and creating them where they did not formerly exist, the emperors main-
out, see my comments in American Historical Review, 76 (1971), 425-6 with nn. 63-6.
3 A useful indicator of the normality of munera is Ulpian's dictum (Dig. 50. 4.4.2):
tained the conditions best designed to ensure willing, or at least uncoerced,
Inopes onera patrimonii ipsa non habendi necessitate non sustinent, corpori autem contributions to imperial needs.
indictaobsequlasolvunt/'* When Diocletians tax reform is set against this backdrop, one may
4 The Urso charter, ch. 98, in FIRASI, 189; the African inscriptions, ibid., pp. 489, 497. see that it did not, at first, represent a major change in the ways by which the
Because the services prescribed in the African inscriptions are agricultural, they have
often been thought to prefigure the labour services of the Carolingian polyptychs (or emperors obtained revenue. Changes there certainly were; they were
P. Ital. 3); in this sense, lately, J. Percival, "Seigneurial Aspects of Late Roman Estate
Management," English Historical Review, 84 (1969), 459-60. The amount of service 6 Cf. above, ch. 11,n. 26 . It is typical that a recent general work should look upon munera
prescribed, however, is so much less that the clausesremind one rather (as Mommsen as a third-century innovation, one of two ideas of the imperial government (the other
observed)ofthepubliclabourprovisionsoftheUrsocharter.Forthelaws, above, n. 3 *. beingcurrency debasement) for remedying its financialplight: G. A.J. Hodgett, A Social
5 Ardant, Hist.deI'impot,i, 32;PhilippeSagnac, LajindeI'Ancienregimeet laRevolution and Economic History of Medieval Europe (London, 1972), p. 37.
americaine(i^j-ljSg), 3rded., Peopleset civilisations 12 (Paris, 1952),pp. 318-21. 7 Cf. Ardant, Hist. de I'impot, I, 55 n. i.
 94 CaputAviv COLONATE CONCLUSION 95

occasioned chiefly by the collapse of the currency, with the consequent on the non-monetary basis of "shares of assessment. " Citizens including
instability of money values and the government's resort to exactions of tenant-farmers had capitatio, "tax liability," indicative of their personal
annona, rations for the army. Continuity prevailed, however, in the em- capacityto contribute to the annona;the "heading"ofa citizen in the tax
perors practice of dealing with the taxpayers through their local units of rolls, embracing the whole ofhis professio censualis, was called his caput; anA
government ratherthan individually. Therole oflocal government is even caput was also the preferred term for the abstract "share of assessment,"
likelyto havebeenmagnifiedbytaxationinkind,forit dependedmorethan upon whose basis tax levies were distributed among the taxpayers pro-
ever on a knowledgeoflocal circumstancesthat imperial officialscouldnot portionately to their wealth. By the close ofthe fourth century, however, we
have had. The new assessmentscheme wasnot intendedto alter the relation- contemplate a system of assessment that was rooted in land; in which all
ship of the two levels of government. The fact of taxation in common shares ofassessment, regardless ofname, were understood to be secured by
commodities meant that the number of payers of direct taxes could be land; where, moreover, coloni were obligated by the state to cultivate the
increased, and the instability of money values made it impossible to apply landin lieu ofother contribution to taxation; andwherecapitatio wasglossed
the traditional quantitative measure to the property declarations on which asrerum tributa and munia agrorum. The share ofassessment called iugum, it
assessment was based. Though these were new conditions, the imperial appears, eventually prevailed over the one called caput.
government hadno compelling reasonto invent uniform institutions to cope The later scheme developed out ofthe earlier one without our being
with them. Local authorities, if sufficiently pressed to maintain or increase able to establish with precision the chronology and mode ofthe process.
their tributary payments, could be expectedto find expedientsappropriate ThecompilersoftheTheodosianCode,in whosetimeassessmentappliedto
to their individualcircumstances;theyhadno doubtbeendoingsoalready. land, continued to be conscious that capitatio meant tax liability in general,
What might cause concern to the emperors was the political equilibrium: and that shares of assessment could bear the name ofcapita as well asiuga.8
unless action of some kind were taken, the fiscal pressure exerted from a They may havebeenin somedoubtaboutthe originsofthesecapita,which
distance by the imperial court was likely to be translated into a continual had been wrongly glossed in 430 as discriptio animarum; but they knew that,
growth ofthe powers oflocal government over their citizens, for it wasonly asmatters nowstood, suchcapitawereasconcreteasiugaanddifferentfrom
just that the collectivities which were called on to contribute more should be capitatiohumanavelammalium.9 Bythenalso,aneasternbishopcouldsuppose
allowed a freer hand in exacting the needed goods and services from the that iugashould ideally be related to the square mileage ofthe territory to
taxpayers.Whatwasjeopardized,therefore, by thenewconditionsofecon- whichtheyapplied;on theonehandstoodtheacreageofthedistrict, on the
omic life and taxation was the balance of power at the local level between other the tax units - excessive in terms of acreage - for which the district
imperialandlocalauthorities.Ifprovincial governorswereto exercisetheir was held accountable. 10When such ideas were voiced, the iugum had come
traditionalsurveillanceandappealsjurisdiction,a newlanguage- other than closeto beingregardedasa term ofsurveyratherthanaswhatit hadorigin-
money - hadto be devisedin whoseterms the actions oflocal government ally been, namely a measure for the tax liability offortunes m land. 11
in tax assessment, apportionment, andcollection might bejudged by imperial What were the capita that the legal compilers of the fifth century
officials.The schemeo£iugaandcapitaservedthis purpose. It wasan emin- understood to be as concrete as iuga? This is the problem that modern com-
ently conservative measure, designed primarily to preserve the supervisory mentators have had the greatest difficulty in solving. The answer I have
role ofthe imperial government over the local authorities, which asalways
were expected to carry out the actual taxing ofindividuals.
8 Cf. Ccrati, Carac.annon., pp. 32-41,for comparableconsiderationswithregardto the
The difficulties commentators have had with the terminology oflate words tributum and annona.
Roman taxation, and consequently with a reconstruction of its forms, stem 9 CTh 11. 20. 6; cf. above, ch. iv, n. 55.
from the fact that the sources document an assessmentscheme in the early io Theodoret ofCyrrhus Epist. 42, ed. andtr. Y. Azcma, Sourceschretiennes98 (Paris,
1964), pp. 109-13.The letter is addressedto the PraetorianPrefectat Constantinople.
fourth century that differs considerably from the one that prevailed at its 11 Although Theodoret thought of the iugum almost as a term of survey, we have no
close. What we witness at the beginningis a system ofpersonalassessment reasonto suppose that the administrationtreated it in this way in the fifth century.
 96 CapU t AND COLONATE
CONCLUSION 97
offered is that these capitawere a residue ofthe earlier assessment process. payers individually responsible for multiple, whole, or fractional shares of
According to that scheme, which was not applied to the whole Empire, assessment, whose total was communicated to the government. But the
taxpayershadbeenassignedsharesofassessmentonthebasisoftwoschedules, elements in the conversion schedules were all subject to the mutability of
one appreciating productive land, another appreciating men and animals, time: men and livestock are born, age, and die; productive land, though
forconversion intoa uniform assessment unit,callediugumorcaputdepending relatively stable, is brought into and drops out ofcultivation. The govera-
onthe schedule it wasbasedon. Unless a professio censualis consisted ofonly ment records enshrined a territory appreciated in stablesharesofassessment,
one ofthese categories ofassets (which could rarely have happened), it was whereas thecommunities to whichtheserecords applied consisted ofmortals
siftedthrough both schedules in order to produce the total ofsharesthatthe and lands subject to mutability. What is more, the rate ofchange between
declarant was accountable for.
thelanc^^ormc)fassessmentshareandtheanimateformdifferedradically.
Thewordusedto expressthetotal ofsharesthusattainedisonepoint Civic spirit had no doubt benefited from the stress laid upon the personal
at which terminological confusion may have occurred from the first as a aspect ofassessment, since this exemplified the character oftaxation asactive
result ofdiversity in local usage. Strictly speaking, every taxpayer declaring functio rather than passive praestatio. But, once the new assessment had been
both land and mortal assets should have been assigned an assessment in instituted, the officials mediating between real taxpayers andabstract assess-
iugavel capita{zugokephalai). l°tBut, sinceiugumandcaputwerebothabstrac- ment shares had good reasons for preferring sluggish mga over fluid capita:
tions andidentical in value asunits ofaccount, there was no practical reason iuga stood still and offered solid security for tax arrears; capita tended to
why totals based on the conversion ofboth types ofassets might not pass evaporate.
under the name ofeither unit. It could hardly be otherwise when capitatio In attempting to reconstruct what happened to iugaand capita as the
was commonly used for assessment consisting ofiuga as well as capita, and fourthcentury woreon onemustavoidexaggeratingtheshortcomings of
when iugatio m a general sense was likely to embrace animate assessment as the assessment scheme. Ifthe system had been grossly defective, its animate
well as land. Uniformity in terminology and records were unimportant component would have collapsed in short order or, more likely, would not
considerations. The objects ofthe reform were to avert unfairness in local havebeeninstituted atall. Whateverthedefectswere, theybecameevident
assessment by imposing a non-monetary accounting unit, and to assist over a course of decades rather than years. As late as 368 and 370, we still
imperial estimates by supplying the central bureaus ofthe government with find the government treating taxation as personalfunctio rather than as a
totals ofsharesperadministrative district. As long astheassessment process payment for land. 13In time, however, the practice ofdual assessment was
hadbeenuniform- thecollection ofdeclarationsoflands,persons,andlive- displaced. The clearest evidence we have for the change is the dismissal of
stock, andtheconversion oftheseprofessionesinto sharesbytheapplication tenant-farmers fi-om the assessment rolls and their assignment to the land.
ofthe two schedules- thespecificterm locally applied to the share-unit did Theeffort tomaintainthemas"active"citizens,contributing totaxationon
not haveto be strictly controlled. the basis of'unsecured shares ofassessment, " eventually broke down.
So far so good. What mattered wasnot uniform terminology but the Though obscure in details, the course ofevents can be grasped in
capacity ofeach community to apportion its tax bill fairly among those tax- outline. Thegovernment, fromitslofty eminence, strove tokeepcommuni-
ties responsible for their totals of shares. Its attitude was not inflexible:
witnessthecaseofAutun. 14Buttheschemeit instituted wasoriented towards
12 As C&atirightly pointed out (Carac. annon., p. 403), this senseofzugokephaleis
expressly attested in the undated Astypaleainscription(see Appendixl). In the sixtU stability in the tax base, andnot towards the frequent adjustment oftotals
century, however - presumably much later than the inscription - the character of the either up or down. On the local level, magistrates tended to focus their
zugokephaleasanexclusivelylandedshareofassessmentmaybeinferredfromCJ10.27.2.8
(491-505) andjustinianNor. 17. 8,whentheirlistsoftermsforsharesarecomparedwith 13 Above, p. 38.
the oneinJust.Nov. 128.1 (545), which,owingto the inclusionofthecenturia,demon- 14 Above, p. 50; for laws excising assessment, above, ch. iv, n. 63. Cf. Eusebius Vita
strably refers to land alone. The confinement ofcensusto land, to the exclusion ofall Constantini4.2-3.Also, in thenextcentury,thecommentofTheodoretaboutneigh-
other assets, is specially stressed in Nor. 168 (512?). bouringcities:Epist.47,cd.Azema(asabove,n. 10),pp. 122-3.
 98 CapU t AND COLONATB CONCLUSION 99

interest on the iugarepresenting productive land, especially when it came to assigned to the soil they cultivated or that land itself, rather than personal
recovering arrears ofpayment. With the capita,they did asbest they could. fortunes, should have become the basis of assessment? The likelihood that it
One constructive step available to them was to discover and specify the did matter is one thing; specifying how it did isanother, which should not be
proportion ofthe capitathat consisted ofreal property; another was to hold attempted except on the basis ofmore extensive source material than has been
substantial landowners responsible for the assessment owed by their tenant- considered here. The present study is no substitute for the exhaustive history
farmers. In these ways, bad debts arising from animate shares ofassessment oflateRomantaxation that will eventually haveto be written. It is appro-
could be minimized, even ifthey could not be altogether avoided. Over the priate nevertheless, before closing, to put forward three points for further
years,suchadministrativepracticesmight bringinto existencea newkind of discussion. These are hypotheses suggested and partially supported by the
zugokephale,not onerepresenting (asoriginally) theadditionofinanimateto evidencejust surveyed but by no means beyond dispute. Even ifthey should
animateassessment,butrathera kephalethathadthecomfortingsolidityofa subsequently be disproved, our knowledge of imperial taxation may profit
zugon.19Communitiesmightcometo regardthemselvesashavingtwo sorts from their being aired.
of shares of assessment: on the one hand, iuga vel capita that were reliable
because backed by land declarations; and, on the other hand, the unsecured I How was provincial land taxed in the early Empire?
capitaofthe old assessment- a censushumanaecapitationisthat wasa liability
rather than an asset, since collections from it entailed greater trouble than its A recent historian oftaxation has pointed out that hardly anything is more
tax yield was worth. When this situation wasreached, it remained only to difficult to tax than agriculture, especiallyagricultureconducted on a small
persuade the government to adjust its total ofassessment shares to the lower scale with a considerable element ofsubsistence production. How is the state
figureandto disposein somewayofthe balance. 16Aftertheadjustmenthad to grasp "la matiere imposable'
taken place, some ofthe old terminology remained, but that which it referred
to had been transformed. What was expected ofthe taxpayers henceforth ... where everyone strove to produce what he needed, where the
was a praestatio for their land. peasant baked his bread, ate the meat and spun the wool ofhis flock,
The evidence analysed in this study permits us to describe the course wove the fabric of his clothes, pressed his grapes, made his furniture
taken by one aspect oftaxation during the fourth century but is too limited and tools: how could the fisc take hold of this abbreviated circuit-
tojustify ambitiousinterpretativeconclusions.Didit matter that coloniwere the shortest ofall economic circuits? Worse yet, the producing units
were small.An army ofinspectors would havebeenneeded to establish
15 As above, n. 12. Cerati, Carac. annon., pp. 403-12, rightly pointed out the disparity
between the usage ofzugokephale at Astypalea and in the sixth-century laws; a community
thenetprofit, andtheywouldhavecomeup againstpeasantsinnocent
assessed in iuga ve\ capita (i.e., zu^okephalai in the sense exemplified by the inscription) ofthe slightest notion ofkeeping accounts.
would automatically have had a higher rating than one assessed in wga alone (i.e., the Finding out the volume ofthe harvest would not have been
unit equivalent to zugokephalai in the laws), and this would have been manifestly unfair. I
suggest that, when all assessment was sifted through two schedules of units equivalent
enough.In theabsenceofexchanges,howcould oneknow its value?
in tax yield, it was a matter of indifference whether the total was expressed in capita How could the costs ofcultivation be establishedwhere one dealtwith
(asatAutun or inCT7»7.4. 32), or inwga(asin suchlawsasCTh 15. 1.34or 16.2. 15),or in family enterprises wheresalarieswereneither fixednor paid, andwhere
zugokephalai== wgavelcapita(asatAstypaleaandin manyLatinlaws). Becausethethird fertilizer, rare in any case, came from the manure ofthe animals?17
term included both units, it was obviously the most precise one, but either capita or iuga
alone could express the same meaning by being used generically, in thesenseof "shares
ofassessment, " which is what they were. Subsequently, however, the changes in assess- Ancient agriculture was unquestionably profitable; it was the mainstay of
ment took place that havejust been sketched. By 430, iugasive capitareferred to landed almost all fortunes, and the principal form ofproductive activity. But know-
assessment only (CTh 11,20.6). It is logical that the Greek equivalent, zugokephahi,
should have changed in the same way and acquired the exclusively landed meaning ing this was different from being able to fix on some specific crossroad at
documented in the sixth-century laws.
16 The outstanding illustration is the law for Thrace, quoted above, pp. 86-7. 17 Ardant, Hist. de I'impot, I, 337, speaking about Europe in the early modern period.
 100 CapU t AND COLONATE CONCLUSION 101

whichthecommunitycouldinterveneto takea shareoftherevenuearising professiones, in a context suggesting thatcivicvirtue residedin makingthem

from land. honestly.22Oneisthereforetemptedto concludethat, whilefieldsburdened
Perhaps the most common charge on land, other than the costs of bytributewereodious,taxationonthebasisofdeclarationswasunobjection-
cultivation, consisted ofrents. Rents were hardly trouble free, though many able. The same contrast between good and bad taxation is suggested by
private fortunes depended upon them. 18 As owners of public lands, some- Aurelius Victor's account of the tax reforms of Diocletian. Under this
times of considerable extent, collectivities themselves relied on such emperor, he tells us, Italy was subjected to the tributoruin mgens malum;
revenues. 19Iftaxeswereassesseddirectly on land, they wouldhavehadthe formerly, allItalyborethesamefunctio offeedingtheimperialcourtandarmy
characterofa rentandwouldoftenhaveweigheduponlandwhoseproduc- that were almost always there; but the new law imposed pensiones, which
tion already had one rent to bear. were tolerable at first but have risen to a ruinous level. 23To put it another
Two literary texts arevery difficult to interpret unless onepresupposes way, the fact of contributing rations, annona, to the court and army was
that taxes in the form ofrents were regarded asexceptional and repugnant. unobjectionable; the mgens malum consisted in Italy's being subjected to
The context of the first is Tertullian's illustration ofhow little the pagans tributa, glossed a little later by the word pensiones, "rents."
revered their gods, and how little reverence the gods deserved; "By public Ifthe distinction these passages appear to make points in the correct
law, he says, you defilethe public [gods] by auctioningthem to conces- direction, then it follows that the inhabitants ofthe Roman Empire, though
sionaires ... Fields burdened by tribute [are regarded as] of less value, the taxed, didnot normally look uponthemselves aspayingthestatesomething
heads of men assessed for payment [are regarded as] less honourable: for comparable to a rent for their lands. Rather, it wascommonplace among them
these are marks ofbondage. With the gods, however, the scale oftribute is that taxation of this kind was pernicious and odious. The provinces, to be
themeasureofholiness:theholiera godis, themoreheyieldsin tribute."20 sure,mightcontaintractslyingoutsideorganizedcommunities - "tributary"
Ifthe point ofthe passage is kept m mind, it becomes apparent that Tertul- land that the imperial government leased for a rent to those who would take
lian s reference to taxed fields and heads is meant to evoke neutral assent it on the terms offered. But where organized communities existed, the pay-
from the audience, and not self-pity. The implication is that pagans do to ment to Rome oftributum (whatever it consisted ofin origin)24 wasa charge
their gods what they would feel defiled to endure in respect to themselves uponthelocalrespublica;andhowthecommunityrecoveredthissum from
and their property. 21 Somewhat later, Tertullian mentions tax declarations, its citizens was a matter of its own public finances, under appropriate but
distant imperial control.
l8 For difficultiesoccasionedby rents (in tribute-freeItaly), seePIinyEplst. 3 ip, 4.14, This brings us backto the difficulties oftaxinga predominantly agri-
9. 15, 36-7. In a specialcase[Epist. 7. 18), Pliny expectsto haveno trouble in finding a cultural economy. Ifthecommunities oftheRoman Empiredidnotimpose
tenantfora property,becausetherenttobepaidonitsaccountwasmodestbycomparison rates upon eachunit ofcultivation in the manner ofa rent, what course did
with its capital value. I take Pliny to document nothing more serious in hisletters than the
ordinary cares andtroubles of (largely) absenteeowners. they adopt for grasping the elusive "matiere imposable" and obtaining the
ip Jones, Greek City, p. 359 n. 67, with reservations about his remarks on land tax. On
thetraditionalmismanagementofmunicipallands,PlinyEpist.7. 18.
20 Apologeticum 13.5-6: "Publicos [deos] aequepublico iure foedatis, quos in hastario 23 Apohg. 42. 9.
vectigales habetis ... Sed enim agri tributo onusti viliores, hominum capita stipendio 23 Liber de Caesaribus 39. 31-2: "Hinc denique parti Italiae invectum tributorum ingens
censaignobiliora,nam haesunt notaecaptivitatis; delvero, qui magistributarii, magis malum. Nam cum omnis eadem functione moderateque ageret, quo exercitus atque
sancti; immo qui magis s.mcti, magis tributarii. " On this work and its audience, T. D. imperator,quisemperautmaximapaneaderant,allpossent,pensionibusinductalexnova.
Barnes, TertulUan: A HistoricalandLiterary Study (Oxford, 1971),pp. 107-10. Quaesane illorum temporum modestiatolerabilis in peraiciem processit his tempesta-
2i Cf.A.M. Andreades,A HistoryofGreekPtiilicFinance,tr. C.N. Brown (Cambridge, tibus. The basis for saying mgens malum does not seem to be confined to the rise in
Mass., 1933), p. 127. Though he and others havecited the passage to document the struc- pensiones; it embraces the forms oftaxation aswell. On the other hand, Victor's explicit
tureofprovincialtaxationinTertullian'stime (landtaxandheadtax),I donot seehow, reference to the present alerts usto the possibility that hischaracterization ofthe system
in context, sucha meaningcouldhavebeenintended. For a parallel to the ager tribute (i.e., pensiones) holds true for his own time, the 3605, rather than for Diocletian's.
onustus, see Pliny Epist. 7. 18, where a property burdened by a perpetual rent charge, 24 To be sure, tributa often succeeded the taxes remitted to the former regime (e. g., the
payable to a municipality, is explicitly stated to have lost value. regia tributa of Cappadocia, above, ch. I, n. 29), and not simply a land rent to Rome.
 102 CapU t AND COLONATE CONCLUSION 103

wherewithalfor meetingtheir obligationsto higherauthority?The solution that the continued flow oftribute was best guaranteed by placing it under
appears to have been to eschew taxation ofproperty as such and to set in its local administration, implicitly renounced the possibility of sharing in the
place a body of public-spirited contributors to communal finances. To be eventual enrichment ofthe provinces. A number ofdo s and don ts limited
sure, public spirit alone was not counted on to replenish local treasuries. the activities oflocal governments in the management oftributes, and pro-
Censuses ofcitizens and their property took place once in a while, and some vincial governors could receive appeals against localities that transgressed
attempt may evenhavebeenmadeto keep them up to date; ifa once-poor these rules. But beyond the general expectation that the rules would be
man built up a fortune, the likelihood that he would be nominated to local observed, the imperial government had no specific knowledge of how the
officesandmunerawasbound,sooneror later,to bringhisfortuneintopublic tribute it received was raised. The security of its revenue resided not in a
view. 25 But, whatever course was taken to keep track of taxable resources system ofverification based on such records as censuses oflocal assets, but in
and to impose a check on the honesty of declarants, the communities did the reliability and solvency ofthe governing bodies that ruled each locality.
not attempt to levy a direct tax on land or agricultural production. On the A central government that relies on intermediaries of this kind is
contrary, thespecialbenefiteachcommunityconferreduponits citizenswas generallyill-informedabouttheresourcesofits subjects:"Thosewhosetask
ratherto liberatetheterritory from theimperial "rent" weighinguponit, if it isto carryouta sub-distributionoftaxationprefer- ingeneral- nottokeep
such had existed, and to leave the land, its owners, and its cultivators entirely a trace ofit... In the caseofan urbanor village collectivity, it prefers not to
free for the many kinds ofprivate affairsthat producedindividualincomes. supply higher authority with the means ofrealizing the possibility ofincreas-
What the community counted upon instead ofa land tax was something ing the [tax] burden."26 In the Roman Empire, the personal character of
far more precious,namelya personalcommitment by thosebenefitingfrom local direct taxation helped to hold at a distance the potential inquisitiveness
land to sustain the finances of the locality. In these citizens, reared in the ofthe imperialgovernment, for even localitiesrelied rather on public spirit
spirit ofpublicmunificence,the communities ofthe Empirehada "matiere than on records to maintain their solvency. One need not even suppose that
imposable"thatmightindeedhaveoffereda practicalsolutiontotheproblem there was a deliberate motive ofexcluding the prying eyes ofhigher author-
ofhow to tax an agricultural economy. ity. Rather, the forms oftaxation, allowing little control ofhow wealth was
acquired,werethosebestdesignedto keepalivethespirit ofgenerositythat,
2 The exclusion ofthe imperial government from local economies. with minimum pressure, kept public revenues flowing.
It is apparent that, with tributes structured in this way, Rome was
The communities of the Empire benefited from the conditions of taxation poorlyplacedto facea riseinimperialexpenditures.Ifprovincialenrichment
undertheearlierEmpire.Theywereresponsibleto theimperialgovernment increased interregional traffic, the proceeds ofthe portoria would no doubt
for a fixedtotal that wassupposedto beattainedby the proceeds ofcertain grow in proportion. Otherwise, however, something other than the fixed
designated taxes, whose administration was in their hands. By setting a limit tributes would have to be devised in order to tap provincial wealth. One
on the obligatory export of local wealth, the fixity of the total offered an proposal of Cassius Dio about imperial revenue, early in the third century,
incentive for the intensification oflocal economic activity. It also meant that wasthat a huge sum ofmoney, realized from thesaleofimperial lands, should
any increase in the wealth generated by the local economy was available belent to landownersin return for a modest, perpetualinterestanduponthe
either for private enrichment or, ifa way existedfor turning it towardsthe pledge of ample property as security - a scheme closely resembling the
publicpurse, for localpublicpurposes.Theimperialgovernment, convinced financingofthealimentaryfoundations.27Whatthisamountedto,apartfrom

25 For some examples of how nomination might marshal fortunes for the public interest, 26 Ardant, Hist. de I'impot, I, 53; the specific casecited (p. 54 n. l) is that the French
see CJ 7.64. 3 (Gordian), where someone is named duumvir while still appealing his clergy under the ancienregime kept from the royal administrationthe details of how its
nominationto thecity council; CJ9.21 (ca. 300),specifyingthat ordinaryfreedmenmust don gratuit" was shared out.
not benameddecurions,is suggestiveofhow, if a locality were financiallyhardpressed, 27 Cassius Dio 52. 28. 3-4. This is the most novel of several proposals. Cf. F. Millar,
nomination might break down social barriers. A Study iifCassills Dio (Oxford, 1964), pp. 109-11.Onhow the alimentaryfoundations
 104 CapU t AND COLONATE
CONCLUSION 10$

the initial loan, was the imposition ofa small, perpetual, fixed rent upon a doret illustrate that the assessment shares ofCyrrhus were not simply the
large proportion ofthe real property ofthe Empire. 28 Since such a scheme single mass they had been in the oration about Autun. Rather, Theodoret
presupposed a stable currency, its adoption would have foreclosed another represents the shares both as a total and as three slices assigned to specific
course of action, which the emperors had already engaged in and were to branches ofthe imperial government - 10,000to theresprivata, 15, 000to the
resort to repeatedly in the decades to come, namely depreciation of the sacraelargitiones,andthebalanceof37,000to thepraetorianprefecture. 29The
coinage. Either Dio's proposal or the course actually taken illustrates the letters illustrate a crucial aspect ofiuga aud capita, namely that they were in
problem ofa stateattempting toexpanditsincomewhenunabletodosoby current use in the calculations of the central and provincial bureaus ofthe
raising tax levels. Though beneficial in many ways, the organized communi- imperial government.
ties ofthe Empire also seta limit on the actions ofthe central government. Shares of assessment had been introduced as a substitute for money
The communities might well show resilience in responding to imperial in measuring taxable assets. 30But wgaand capitawere different from money
demands; but the advantages afforded to the emperors by resourceful cor- andpermitted calculations ofa kind that had beeninconceivable whenmoney
porations ofthis kind appear to have been incompatible, at least in the earlier had been the unit ofaccount. Money had been merely a measure, a way to
Empire, with raisingthe proceeds ofprovincial tributes on a rational basis. distinguish quantitarively between the assets of one taxpayer and those of
the others. The referent of money in the older form ofprofessio censualis
3 The effect ofiuga and capita upon the integrity oflocal economies. had been the local community. One needed to know a citizen's census in
orderto determinenotsimplythedirecttaxesheshoaldpaybutratherwhat
When one focuses, as in this study, on the terminology ofiuga and capita he should contribute to the community in a variety of ways - periodic
inthefourthcentury, nomajorchangeneedbesupposedtohavetakenpiace liturgies and unpaid offices, as well asmoney or goods. On the other hand,
intherelationsoftheimperialgovernmentwiththetaxpayers.Localgovern- this census stated in money had had no specific relevance to the imperial
ments continueto look like theindispensableintermediaries,andtheirtreat- government. The conversion to money was a pure abstraction, stating a
ment ofsharesofassessmentmayberepresented asthe guidingthread ofthe man's worth where he was a citizen or resident; it was only there that his
movement awayfrom personalfortunes and towardsland. To someextent census was convertible into private wealth and public functiones; and, as a
thisimpression iscorrect. Intheletters where Bishop Theodoret complains of result, it was only through the community that the imperial government
the number of shares weighing upon Cyrrhus (446-7), the most visible might drawrevenue (directly) from a taxpayer'sresources.31
contrast with earlier evidence, such as the Panegyric about the taxes of Likemoney, liigaandcapitawere also anabstract quantitative measure
Autun, is the rank of the speaker rather than the nature of the relations ofassets, distinguishing between the varying tax liability ofindividual tax-
between the locality and higher authority. Nevertheless, one may wonder payers. The language ofshares ofassessment appears to have been originally
whether the relationship had not in fact changed. Even the letters ofTheo- devised to do only this. But mgaand capita were more concrete than money.

were organized, P. Veyne, "La table des Ligures Baebiani et 1'institution alimentaire de 29 The statistics offered by Theodoret are not straightforward.*
Trajan, Ecole frangaise de Rome, Melanges d'archeoloye et d'hisfoire, 69 (1957), 125-35, 30 It should berecalled that the key text documenting property declarationsin the earlier
70 (l958), 222-41.
Empireis Dig.50. 15.4, whereUlpianspecifiesthateverythingdeclaredmust beestimated
28 N.G. Svoronos hasproposed tbat, towards lioo, Byzantium hadone tax rate on land in money. In an economy structured like ours, it is normally expected that property can,
as capitalstatique," estimatedby himat about 1.33 to 2% ofgrossreturn, andanother at almost any time, be converted into money, and probably for more than its declared
on la terre travaillee, doneessentiellementsur la production," estimatedat6 to 10%: value. It is doubtful that Antiquity could have had the same expectations; "book value"
'Recherches sur Ie cadastre byzantin et la fiscalite aux xre et xne siecles: Ie cadastre de did not necessarilyimply the parallel existenceof "marketvalue."
Thebes," Bulletin decorrespondancehellenique, 83 (1959), 141.Dio'sschemereminds one 31 AsidefromtheanecdoteaboutCaligula(above,ch.i, n. 67),whichinanycaserefers
ofthe Byzantine tax on land as"capital statique. " The loan aspectofthe scheme is by to persons,noevidencesuggeststhattheimperialgovernmentrecordedthemoneytotals
nomeansalientotaxation; largeloanstoimportant menmarkedthebeginningsofRoman of the censuses of individual communities or provinces. Doing so would have been
taxation in Britain after the conquest: Cassius Dio 62. 2. 1. pointless.
 106 CapUt AND COLONATE CONCLUSION 107

They not only measured but actually referred to concretely productive land, have supposed that the first emperors caused a Reichscensus to be compiled,
men, andanimals; andtheir specificcharacter in taxation wasprecisely that a view that now seems highly unlikely. Although the fourth-century totals
eachone,intheory, yieldedtheannonariapraestatiolaiddownpershareinthe ofshares ofassessmenthardly constituted a Reichscensus,they cameas close
year s delegatio. They were more transparent, therefore, than money had been to beingoneasanyadministrativerecordthathadeverbeengatheredbefore.
and might lead to the supposition that higher levels ofgovernment, notjust Here wasa schematic survey ofthe resources ofthe Roman world, reducing
the local one, might collect what the shares owed. It would have been absurd to simple units a productive capital of lands and men that might be relied
to dothiswithassessmentsexpressedinmoney; formoney merely measured upon to bearthe annualcosts ofempire.
value and did not reproduce itselfunless specifically applied to this purpose, The most striking trait ofsuch a record was that it masked the indivi-
such as by being lent out at interest. 32 luga and capita, however, lacked the duality of cities and the heterogeneity of local economies. It portrayed a
intangibility of money; it was perfectly feasible to cross such shares off a homogeneous territorial state, partible in whatever way one wished regard-
local assessment roll and assign them rather to another agent ofcollection, less ofhistorical circumstances, and organized for the primordial purpose of
such as, in a documented case, the household ofan imperial princess. 33 In balancing imperial expenses. Thechangeinallthisfor theorganization ofthe
time, the iugum was to figure explicitly in bureaucratic usage asa "share of Empirewasprobablytoogreattobeimmediatelynoticed,evenbya govern-
tax proceeds ; and much earlier than this the shareofassessmentlent itself ment inclined to centralization. Local units and their officials retained im-
to being disposed ofby theimperial government in sucha wayasto interfere portant roles in the tax process. But a world of shares of assessment lent
with, and perhaps to tear apart, the integrity of local communities as units itself, asoneofcitizen-taxpayershadnot, to generalregulationsissuedfrom
ofpubliceconomy.34 the seat ofcentral government.
We do not know in what form the tributa ofthe provinces had been Thesegeneralrules, substitutingempire-widenorms for theactivities
tallied inAugustus's Breviarium imperil. Theymaywellhaveappeared astotal of local authorities, may well be a greater innovation in fourth-century
proceeds, not even broken downin greater detailthanbyprovince. 38Inthe taxation than the language ofiugaand capita itsel£ One law denied localities
fourth century, the data available to the imperial bureaus were far more therightto granttaxreliefto individuals.36Anotherwithdrewthedistribu-
specific,sincetheyconsistedoftotalsofinterchangeableunits,iugavelcapita, tion ofextraordinary assessments from the communities and assigned it to
easilyseparablebyprefecture, diocese,province, orlocality. Theshareswere provincial governors, only to transfer this function a fewyears later to a yet
highly abstract in composition, but from the standpoint ofpayment each higher rung in the administrative ladder, the praetorian prefecture.37 Such
one seemed capable of supplying the government with the commodities measures expressed the confidence ofthe central government in its capacity
called for or their money equivalents. Modern writers on Roman taxation to be more impartial towards taxpayers than provincial and municipal
magistrates;thefurtheronestoodfromindividuals,thelesslikelyonewasto
abuse them by acceptiopersonarum.ss On the other hand, the remoteness of
32 Cf. W. Stark, TheContamedEconomy,AquinasPapersno. 26 (n.p., 1956),p. 14: the central government also meant that more formal procedures for keeping
...to ustheideaofmoneyhasmergedto suchanextentwiththeideaofrealcapital
thatwecannolongerdistinguishthetwounlesswemakea specialeffort"(addingthat taxable resources in view had to be imposed upon the payers than had been
thetwoideaswerebynomeansmergedinthethirteenthcentury).It seemsimportant needed in a communal context. Imperial laws therefore prescribed that
to recognize that, while late Roman iugaand capitaresembled money asmere measures privatepropertywasnotto besoldapartfromitstaxliability,thattaxpayers
oftaxableassets,theyalsohada traitthatmoney(testeStark)wasnottohavefora long
time to come: in the eyes of the government, these shares of assessment were units of
productive capital. 36 CTh 12. 1. 17 (329). Cf. CTh 12. 1. 1 (326s), by which the right to grant relief to a
33 CTh 10.25. 1 (406). For the acquisition ofsuch revenues by others, NVaknt 10. 1. 1. curial was deniedto provincial governors andreserved for the emperor.
34 Cassiodorus Vanae4.20, cf. 2. 37; CTh 13. 5. 14 (371); seethe discussion in Goffart, 37 CTh 11. 13. 3 (32js), distribution by localities with the consiUum ofprovincial gover-
"From Rom. Taxat. to Med. Seign.," pp. 168-75. nors; 11. 16.4 (328), by governors alone; 11. 16. 8 (3565), only by the PraetorianPrefect.
35 Above, pp. 7 and 13. 3 8 As indicated above, ch. v, n. 9.
 108 CapU t AND COLONATB
CONCLUSION lOp

werenot totakeuptenancies outside thedistricts where they wereregistered, owner who bought additionalslaves, enlargedhis herds, or planted barren
and that assessed slaves must not bemoved out ofnarrowly fixed limits. 39 hillsides in vines or olives would quickly have burdened himself and his
Suchsalesandmovements ofpersons could not havebeenrecent novelties. community with a higher liability to imperial taxation. Owing to the un-
At anearlierdate,theinconveniencestheyposedtotaxcollectionhadseemed certaintyor modestyofprofit to beexpectedfrom investment in agriculture,
to be outweighed by the advantages they allowed to private business. Now, the prospect of adding to ones tax liability would alone have sufficed to
however, thecentral government wasanimmediately interested party; it felt discouragesuch enterprise.41The incentive, in sucha situation, wouldrather
the inconveniences whenever sharesofassessment did not yield their annual have been in the opposite direction, to eliminate from one's assets every
dividends, but owing to its remoteness from the arena where individual element whose return did not amply offset the liability it occasioned; for
incomes were earned, it wasunable to seethe advantages ofnot obstructing example, once the production of a low-yield vineyard was matched to the
private transactions. iugatioit represented,theadvantageouscoursewouldhavebeento returnthe
The ultimate temptation offered to the central government by shares land to low-quality arable or wasterather than to continue growingvines.
of assessment was that of direct collection from individuals by its own Calculations of this kind depended largely on the question whether tax
agencies - the provincial government or even a palatine ministry. Unpaid authorities reacted quickly to changes in individual and collective assets. If,
intermediaries were useful, no doubt, in collecting from certain categories as seems more likely, totals ofshares remained fixed for a long time both for
ofpayers, especially since the intermediaries formed colleges that could be communities and for the individuals in them, then the tendency to disinvest
held collectively responsible for defectors. But shares were simply shares, need not have been acute.
uniform for wide territories, having no indissoluble relationship to the Nevertheless, the direct relationship between local assessment and
community in which they were originally declared. The laws do not suggest imperial taxation created by the language ofiuga and capita was a novelty
that the imperial government deliberately took upon itselfthe task ofmatch- that hadnever before entered into the calculations ofthe provincials. Imperial
ingabstractshareswithlivepayers;rather,theinitiativecamefromtheother exactionshadformerly comeuponthem from afar,asannualtributearising
side,intheformofindividualsseekingtosimplifytheirpaymentsortodefend from distant historical circumstances and asarbitrary exactions that could not
their privileges. Shares ofassessment were not in themselves a way for the bedenied.Facedwithsuchcharges,communitiesmightlamenttheirpoverty
imperial government to taxindividuals, butthey could serve this purpose, as or take pride in their wealth; but they had no reason to suppose that the
theearlierformofassessmentcouldnot,providedindividualtaxpayersfound burdenitselfwasmathematicallyproportionedto theirresources.42Keeping
a wayto interest the government in doing so.40 track of these resources, organizing the community to face its regular and
Shares invited the imperial government to indulge its bent towards extraordinary expenses, wasa local concern, not wholly hidden from imperial
administrative centralization; they may also have encouraged localities and view or exempt from regulatory legislation, but a communal process all the
their citizens to put on a faceofpoverty rather thanwealth. Wewould have same. While exactions by the Empire were unavoidable, there was no per-
to know more than we do about fourth-century assessment practices, and ceptible advantage in being less rather than more capable ofsustaining them.
especially about the periodicity ofrevisions, in order to establishwithpre- A large and growing assessment roll meant greater ease in paying the fixed
cision the effects otwga andcapitaupon themanagement ofprivate fortunes.
Iflocal and imperial officials had closely supervised assessments, any land-
41 Cf. Ardant, Hist.deI'impSt,l, 374-90,esp. 387-90.Theimperialgovernmentwaswell
aware of the relationship of taxation to land development (see, e. g., CTh 5. 13. 4). One
39 Landsales, CTh 11. 3. 1 (3135), 11. 3.2 (327), 3. 1.2 (337?) == Fragm. Vatic. 35 (342?s), gets the impression that its eagerness to keep land in cultivation eventually degraded the
CTh 11.3. 3 (363); see also above, ch. iv, n. 61. Movement of tenant-taxpayers, CTh meaning ofthe term emphyteusis from a contract to improve to one simply denoting to
5-I7-I (332). Slavesnot to be sold, outside province, CTh 11. 3.2 (327), cf. 7. 1. 3 (349); cultivate.
apart from fields, CJ 11.48.7 (371). 42 A calculation of this kind may well have taken place when the fixed tribute was
40 E.g., Cn 11.4. 1 (372),8. I. I2(382),11.7. 12(383),11.22.3 (387),6.3.3 (396),11.22.4-5 established or adjusted (e. g., by Vespasian, above, ch. I, n. 42); but this was not a continu-
(409-10). mg process, repeated with regularity.
 110 CapU t AND COIONATB
CONCLUSION III

tribute, without implying that, as a result, extraordinary imperial levies The contention underlying this study is that the sources about late Roman
would be larger or more frequent. Imperial exactions were one thing, local taxationdocumentmuchmoreextensiveandfar-reachingchangesthanhave
assessmentanother; sinceno administrativelink existedbetweenthem, there beenhithertosupposed.Onewaytorealizethisisto refineournotionsabout
wasno reason to suppose that the one would rise or fall in direct proportion taxation in the earlier Empire; another is to take stock ofthe language of
with the other. On the contrary, communities had always been able to sus- iuga.and capita and to assessits effects upon the communities ofthe Empire.
tain the hope that the growth oflocal assessment would diminish the burden Though the apparent intention ofDiocletian's tax reforms was not to cen-
of imperial taxes. In the later Empire, however, following the institution of tralize taxation, a centralization is what at length took place. When did the
shares of assessment recorded by the imperial government as well as the centre of gravity in taxation slip from the localities to the central govern-
localities, theseconditions cameto anend;asSalvian said, "Our taxesbuy ment? What consequences did the transfer have?
us the privilege ofpaying them forever. "48Not at once perhaps but sooner or We are still far from being able to answer such ambitious questions.
later, the community whose resources had grown would have to pay more, Essential series oflaws- those relating to extraordinary taxation, to privileges,
while the one whose assessment had shrunk would be correspondingly and to imperial lands, among others - require special study before any sort
relieved. Whichofthesefateswasmore desirable?Formerly, therehadbeen ofsynthesis is attempted. Besides, the inquiry must be carried into the fifth
a clear answer to thisfrom the standpoint ofthe community. In a world of century and the sixth, asfar asthe lawsandassociateddocuments extend. In
iugaandcapita, theanswerhadbecome a private matter: with skill, influence, 445> the government ofValenrinianm would havemade a profit if it had
or ingenuity, individuals might find ways to dissociate the apparently firm handedovertheentireprovinceofNumidia,populationandall, to someone
links between taxable resources and tax liability. 44 For communities, how- promising to pay the treasury the bargain sum of 150 pounds of gold per
ever,therewasnoescape;thesystemwassostructuredastoreplacehopeby year. 45 Not long afterwards, merchants were indispensable gears in the
indifference.
machinery of tax collection, and later still a subject of King Athalaric en-
These discussions have been addressed, first, to the question how joyed such extravagant dominium over his lands as to exclude all agents of
agriculturaleconomiesweretaxedintheearlyEmpire;nexttoa reasonwhy public authority andto conduct a private foreign policy. 46In thesephenom-
the imperial government may have faced financial difficulties from the close ena and others, we can dimly perceive the strange effects produced by a
of the second century onwards; and, last, to the economic problems that central government's attempting all atonce, by its soleefforts, to tax directly,
may have been created by the new language oiiugaand capita. How tenta- to honour private property, andto promote the general welfare. It will take
lively they have been put forward must again be stressed. They are possi- much more careful study to dispel the shadows and bring clarity to the pro-
bilities, or lines ofinquiry, inadequately documented and, in the caseofthe cess ofchange.
first one, flying in the face ofreceived opinion regarding solum tributarium. Within a narrower compass, the history of the technical terms of
They invite refutation rather than assent. The purpose of advancing them assessment hasshownthat, inthecourseofthefourth century, theassignment
will have been served ifthey stimulate further discussion, for each oneposes of tax liability shifted from personal fortunes to land, and that the establish-
a problem that deserves to be explored rather than overlooked. ment by the imperial government of objective, non-monetary criteria for
assessment played a vital part in this process. Both iuga and capita were
43 A free rendition ofthe second sentence of Deguhernatione De! 6. 18:' Similiter enim
nos numquam ab hac sumus liberi functione, quam pendemus: ad hocquippe mercecies 45 NVaknt 13. 1. pr.
iugitersolvimus, ut sinecessationesolvamus. " A more literaltranslation: "To thisenddo 46 Merchants:VSValent-ia,(447); NMa/7. 1. 17 (458); EilictumTheodorici149,in FIRA1n,
we incessantlypayout taxesthat theremight beno endto our payingthem." 709; Cassiodorus Variae 2.26, 30, 38, 5. 35; Justinian Sanct. prag. ail petit. Vigil. 18, 20.
44 What is referred to here is not principally tax evasion or even such condemned The role ofnegotiatores may stem from the fourth-century practice by which recipients of
practicesastheinvocationofpatrocinium,butratherthetaxsheltersofferedby theim- multiple annonae commuted them by bargaining withthe keepers ofstored taxcommod-
perial government itselfto categories ofpersons that it privileged and,notably, to lessees ities {apodektai), asindicated by Liebeschtitz, "Money Economy, " pp. 254-5. Bxtravagant
ofimperialland.Thesubjectdeservesseparatestudy. domimum:Varlae 8.23; ProcopiusWars 5. 3.2-4, 9.
 112 CapU t AND COLONATB

personalsharesoftaxation, entirely compatiblewith thetraditionalmunicipal

context oftaxassignment andcollection. But they alsorelated taxliability
directly to the principal resources ofeach community, namely its lands and
men, rather than, asformerly, to fortunes expressed in money. In this
they invited the invasion ofeachlocality by the solicitude ofimperial laws Appendix!
thatcurtailedtheprerogatives ofprivatebusinesswherevertheyappearedto
jeopardize the receipts of taxation. Whether or not these laws were wise, it
can be observed that they did nothing to enhance the role oflocal authorities
or to encourage city councils to take such measures with their own citizens
aswerenecessaryto sustainthelevel ofpubliccontributions. Everymeasure
by-passing local initiative hastened the day when the assignment of tax
liabilitywasdetermined notbyfalliblecitycouncils, butbyobjectiveassess- THE CENSUS INSCRIPTIONS
ment lists to which the imperial government had equal access with local
authorities. Lessthan a century after the new tax system wasinstituted, the A set ofundatedinscriptionsfrom AsiaMinor andthe lonian islandsforms
emperors werea shortstepawayfrom a taskthattheir predecessors hadlong a specialcategoryofevidenceaboutfourth-century taxation.Likethepapyri,
preferred to avoid, namely the directtaxationofindividualcitizens.
thesestoneshavetheadvantage ofbeing "documents delapratique" enabling
Within the limits of its subject matter, this study has presented the usto look beyondthegeneralities ofthelaws. Ontheother hand, commenta-
skeletonofa historyoftaxationinthefourthcentury, a workinghypothesis tors run therisk ofinferring too much from theseinscriptions. They have
needing to be filled out, amended, and refined. Methodologically, it has never been collected asa group or systematically analysed, and there is no
shown that the laws document a succession ofchanges, and arenot simply agreed-upon body ofdoctrine surrounding them; on the contrary, they are
fragments ofa unified mosaic illustrating 250years ofstability. Firm con- usually discussed within the context of a wider argument that tends to
clusions still lie far in the distance; much more work will have to be done, determine those of their features that are stressed.
by many hands, and with a great deal ofmutual correction. Nevertheless, Before turning to the inscriptions from the East, a word should be
there is hope that, in time, the Codes will yield a chronological portrayal saidabouttheonly comparabledocumentfromanotherregion, namelythe
of the socio-economic course of late Antiquity. Volceiinscription of323 {Corpus InscriptionumLatinarum, x, 407). Opinions
are divided over the question whether or not it is similar, the crux of the
matter being whether the abbreviation m should be resolved as millena. a.
south Italian tax unit otherwise attested only from 440 onwards. 1 Deldage
didnothesitate:"Laresolutionde1'abreviation ...estcertaine,etantdonneIe
rapprochement de ce texte et du texte de la Pragmatique concernant une
region voisine. 2 The reasoning is shaky because Justinian's Pragmatic
Sanction dates from two centuries later and, more important, because a
very standard Italian land measure was the modius, that is, the amount of

i NValent 5. 1.4; alsoin NMaj 7. 1. 16 (458), andCassiodorus Variae2. 37,wheremi/fenais

ashareofproceedsratherthanofassessment(above, ch.vi,n. 34);alsobelow,n.3.Inthe
nineteenthcentury,themilknagaveriseto severaltheoriesthatweredisprovedwhenit
wasrecognizedasa unitcomparabletotheiugum;seeLot,"Etendueetvaleur, "pp.14-15.
2 Capitation, p. 223. He also thought he could determine its size.*
 114 CapU t AND COLONATB APPENDIX I: THB CENSUS INSCRIPTIONS 115

land that could be sown with a modius of seed. 3 Owing to this uncertainty, forms of declaration would have been used in the same stone. 7 The various
theinscriptionisoflittle use. Somuchofit islost that we cannotevenaffirm fragments are presumed to include the components of the declarations of
that it is related to the census or taxation.4 individual owners; the article headings, giving the owners' names, are lost.
Although a detailed and, particularly, an objective analysis of the The properties, when addedup, would have been considerable.
easterninscriptionswould be valuable, I do not havethe necessarytraining 2) Mylasa: Deleage, Capitation, pp. 170-2 (nine fragments):8 The very
to conduct it. The descriptions that follow present the observations I made poor condition of preservation makes it difficult to decide whether the
while verifying the discussions ofDeleage andJones. The order ofpresenta- inscription records owners declarations or an inventory ofplanted lands in
tion places first those inscriptions in which fiscalunits - iugaor capita- are not various sections ofthe territory. Arable, meadow, and vine are mentioned,
used.6 with their surface indicated by ordinary measurements. Persons are not
i) Lesbos:InscriptionesGraecae,xii2,nos.76-80(fivefragments).Declarations declared.
of land in ordinary measures; declarations of slaves and livestock in whole 3) Hypaepa: ]. Keil and A. van Premerstein, Bericht iiber eine dritte Reise in
numbers. A scattering ofproper names, ofwhich two are followed by an Lydien, Denkschriften der kais. Akad. der Wiss. (zu Wien), philol. -hist. Cl.,
abbreviated form ofgeorgos, that is, colonus: these may indicate the person 57 (1914), nos. 85-7 (three fragments);Deleage,Capitation,pp. 164-6 (nos.
declaring,onthemodeloftheCosfragment,ratherthananowner'sdeclara- 85, 86). According toJones, nos. 85-6record declarations relating only to the
tion ofcoloni on his land. It seems clear that, in no. 76, col. e, lines 6-10, the village ofDideiphyta (as part ofthe territory ofHypaepa), including persons
propernamesareenteredasownersordeclarantsofthelivestockinquestion.6 without land or with tiny holdings (no. 85), and non-residents owning land
The entries for slaves and other residents lack the details found in the Thera in the village (no. 86). 9 Land is declared in ordinary measures; persons are
and Hypaepa inscript ions. In one fragment only (no. 79), land is listed in two entered in great detail, with their age, sex, and place in the household, and
categories each of vine, arable, and olive trees. There is, as far as I can see, including infant children; a little livestock is declared in no. 85. A noa-
no reason to suppose that this fragment does not form a part of the same resident declares a dependent relative living in Dideiphyta. Each entry of
inscription, though Deleage and Jones both implied otherwise; different no. 85 (illegible in no. 86) closeswith a line reading "total" followed pre-
sumablyby a number. In the only suchtotal wherethe numberis legible, a
3 JustinianSanct.pragm.adpetit. Vigil.26 (554); evenifonerefersto Valentinianm, there man oftwenty living in his own house without land totals one; but the
is a century between the inscription and a referenceto the millena. For modius as a land abbreviation k(ephale) does not appear. 10 The only entry mentioning fiscal
measure, TLL vm, 1241, lines 22-5; 1242, lines 60-4.
4 Deleage,Ca^itadon,p.221,statedthatthelastelementintheheadingwas lasuscription units is no. 87, giving totals in iuga (numbers lost) for the two villages of
du recenseur, Turcius"; but Turcius's quality or functions are not stated. One reason to Cinamura and Dideiphyta."
doubt the relevance ofthe inscription to taxation is that the names ofthe owners are never 4) Them: Inscriptiones Graecae, xm3, nos. 343-9 (seven fragments). One of
given. An Athenian inscription of the second century [[nscriptiones graecae, n-m cd. min.,
part 2, no. 2776) was thought by Marquardt to be related to the census, but Mommsen, in
the richest ofthe inscriptions. Here, demonstrably, the entriesrepresentthe
Hermes, 5 (1870), 129,pointedout that it ismerely anobligatiopraediorum.Thesamemay declarationsofindividualownersandincludeland, slaves, andparoikoi (the
be true of the Volcei inscription. inquilini of Latin texts). 12 Land is declared as arable, vine, and olive trees,
5 The comments ofA.H.M.Jonesare scatteredin CensusRecordsofthe LaterRoman
Empire," Journalof Roman Studies,43 (1953), 49-57; unlikeDeleage,hedidnot conduct 7 Deleage,Capitation,p. 179, appartientpeut-etrea uneautreepoquedu Bas-Empire";
a systematic survey. For a full listing of the editions of these incsriptions, Delcage, Jones, Records, p. 49 n. I. The inscriptionsofAstypaleaandThera are also internally
Capitation, pp. 164-94. This Appendix was completed before I saw the commentary of inconsistent, though not to such a degree.
Cerati, Carac. annon., pp. 385-412,whoseorder of presentation is similar to, andsome- 8 Jones, "Records," p. 49, took no account of it.
what more logical than, mine. Cerati devoted most of his attention to the Astypalea 9 Ibid., p. 55; Deleage, Capitation, p. 167-8, had a different view.
inscription(my no. 7), whichhesoughtto underminesinceherightlysawit assupporting 10 "Records, p. 50;Jonesassumedthat a caputwasmeant.
the "these bivalente, " i. e., that iuga and capita were components of a single scheme. ll Deleagetook no account ofthisfragment- becauseofthe referenceto wga>.
6 Jones, "Records, " p. 51 n. 22. 12 On inquilini, above, ch. w, n. 4*.
 Il6 CapU t AND COLONATE APPENDIX I: THE CENSUS INSCRIPTIONS 117

estimated in ordinary measures. Persons are declared in one only of three thesecondarticleonthefragment.Inthecircumstances,Deleage'sinterpreta-
articles in no. 343, and on no. 346, which has lost the land declaration and tion thatthefirst conversion sums up thethree entries beforeit, andthe second
slaves, if any. Livestock is declared in the second article of no. 343 (a very the entry before (or even analternative, that the first conversion relates only
small holding, without persons), after the persons in no. 346. The third to the line on which it stands), could be entertained;15similarly, the third
entry ofno. 343 (a fairly large holding) lists slaves by name andage, followed mightrefer onlyto a part ofthethirdarticle. On suchinterpretations, arith-
by paroikoi, with the name and age ofall the family, including infants. No. metic produces no satisfactory or consistent results, but in the light ofthe
346 is similar to the end ofno. 343 and very probably lists paroikoi ofanother Syrian tableJones's results are no more convincing; their only merit is to
owner. 13 work out within a certain margin oferror and emendation. Caution advises
So far there has been no question offiscal units. The only place where that onerest content with more modest conclusions. It is clear that, in a part
they appear is in no. 343, twice under the first article, and a third time in the of this inscription, a conversion has taken place from ordinary into fiscal
margin opposite the third article. Jones read the following: i) "they make measures. For lack ofthe official schedules ofequivalents, we cannot expect
i^-^ ^g KZ," 2) "it makes^ ^ 3^ KZ," 3) "S^J<:Z." Thissort ofcon- to unscramble these numbers. 16
version ofordinary measures into fiscal shares ofassessment is what the Syro- 5) Cos: InscriptionesGraecaeadres Romanaspertinentes,iv, no. 1083.Declara-
Roman Lawbook leads us to expect. Disposed as they are, and unique in the tions ofland only, stated in wga, including very small fractions. The inscrip-
inscription, they posea considerable problem ofinterpretation, solved in radi- tion is taken to be part ofthe article of a single owner. Rather than being
callydifferentwaysbyDel^age(whomissedthethird conversion)andJones. designated by name, several plots are stated to be "declared by so-and-so,"
OnJones s reading, the three conversions refer to the three articles ofno. 343, that is, a person named, presumably different from the owner. Deleage
the second being listed in the first article in order to take advantageofits suggested that the declarant might be the tenant. 17
blank space. Each one, then, would stand for the totality ofobjects declared - 6) Chios: Deleage, Capitation, pp. 183-4 (two fragments). Declarations ot
land, persons, and livestock - converted into fiscal measures of K^ephah')- land, inquilini (paroikoi), slaves, and livestock. Each stone is presumed to
Z[uga), whichJonestranslated aswgavel capita.u The trouble is that, even contain the declarations byanindividual owner, though inthiscasethepoint
granting Jones s arithmetic and his emendation ofa figure on the stone, the is open to doubt; the whole stone may refer only to one very large owner.
conversion table consequently arrived at for ordinary into fiscal measures In frag. A, the declarations ofplots are made in zuga and paroikikon kephalai;
works out to roughly double the one laid down in the Syro-Roman Law- according to Deleage, the terminology used implies that urban land is in
book. While the attempt is commendable, the result is so conjectural as to question. Frag. B lists larger holdings, with declarations in zuga, paroikikon
leave in doubt even the premise that the second conversion in fact refers to kephalai, doulon kephalai, and zoon kaphalai. These places were necessarily
extra muros but, Deleage suggested, near the city. 18 The outstanding trait
13 Jones, Records, p. 52, pointed out that the owner mentioned in no. 345 was some- of the fragments is that no numbers are entered; the inscription was dis-
how connected to a Roman senator.
14 Ibid., pp. 49-50; I would agree with him if it is assumed that KZ is identical to the
carded before being completed. 19
Astypalea Z(ifgo)K(ephale), which sbcth-century laws attest to be an ofBcial tax unit (as
above, ch. VI, n. 12). Cerati, Cane. annon. p. 392, stressed that entries without persons l$ Deleage, Capitation, pp. 181-2.
were converted, and judged that Jones s interpretation was perhaps a little hasty. " A 16 The ratio of slaves to free cultivators deduced from this inscription by Jones,
possible alternative, if one conjectures an early fourth-century date, is that K(ephah)- "Records," pp. 56-7 (cf. LRE, p. 793), is based on obscure and unreliable calculations.
Z(ugon) is an informal coinage, in which the prefix K specifies that the Z(ugon) in question 17 Capitation,p. 187;the number 1053 is mistakenlygivenfor thisinscriptionon p. 186.
is a tax unit - this on the understandingthat caputwas the original generic term in the i8 [bid., pp. 184-5.
scheme, as implied by its use at Autun and in Constantine's letter to Arius (ch. rv, n. 38). 19 Ibid., p. 184. Deleage found this particularly interesting and inferred that the frag-
Cf. the much later and very fleeting coinage annonocaput, above, ch. m, n. 14*. Cerati ments nous montrent pour ainsi dire les employes du cadastrepreparant leur travail."
was right to maintain that the Thera inscription, unlike that of Astypalea, does not really I donotquiteunderstandwhatthismeans.Myconclusionisratherthattheideaofaffixing
prove the addition of capita to iuga. the census on stone was given up.
 Il8 CapU t AND COLONATB APPBNDIX I: THE CENSUS INSCRIPTIONS Iip

7) Astypalaea: Insmptiones Graecae, xii3, nos. 180-2, revised in ibid. xn3, [the declaration of] so-and-so, Zu(ga) number, K{ephalai) number.
SuppL,p. 278(threefragments).20Thefragmentscontainthearticlesdevoted Naturally some localities are subject to declarations by several individuals.
to individual owners; no. 180isthemore extensive ofthetwo fragments that Fiscalunits areusedthroughout; capita, however, donot occur in all declara-
containdetails.IfoneadoptstheemendationsofDeleage,no. 180continues tions. 23The declarants include many men ofquality, local decurions, absen-
after the name ofthe owner with a total ofzugokephalai,anda breakdown tees from nearby cities, a Roman senator, and tribunes.
into zugaoflandandkephalai ofmen and beasts. Then follows a list ofplace 9) Tralles: A. Fontrier, in Bulletin de conespondance hellenique, 4 (i88o),
names ofproperties, after each ofwhich is listed (at fullest): the individual 336-8 (one fragment). Eacharticle isheadedby a proper name, followed by
total ofzugokephalai, the zuga ofland, the kephalai ofmen, and the kephalai a title (one priest, three decurions). 24 The articles are made up as follows.
oflivestock. Thelargenumber offractionsusedin all talliesensuresthatthe A place name, slave and animal K(ephalai}, with no details ofnanies and
computation is in fiscalunits. Although Deleage'sinterpretation is sensible, ages for the slaves and no breakdown for livestock; alternatively, a place
and a different one is difficult to conceive, he was nevertheless forced to name, plus Zug(a) with number, K{ephalai) with number. At the end, a
emend all four places where the abbreviation for "heads of animals" is symbol indicates "total, " and this is followed by the total ofiuga only. In
supposed to appear. 21Provided Deleage's emendation is correct, everything otherwords,"heads"arelistedintwoways:inentries without iugaas "heads
in this article has been converted into fiscal measures, and iugaare added to ofslavesandanimals";in entries with iuga,only as"heads. " Neither form of
capitain a ratio ofoneto one. What isnot proved, however, is that, whenthe "head"entrycountsfor the total, whichsumsup iugaalone.25
original conversion took place, one human head equalled a fiscal caput; in
other words, one cannot conclude from this inscription that a human caput These inscriptions hardly speak for themselves. The comments to follow are
wasestimated asa shareofassessment equal to onewgurnofland. In fact, the addressed chiefly to the points made in the body ofmy study.
reverse is proved since the figures following the abbreviation for "human a) The assessment of one adult male as one fiscal caput is not proved. In
heads consist ofcomplex fractions. 22 Hypaepa no. 85, wherethe number "one" isvisible, no reference is madeto
Neither DeleagenorJonespaidattention to no. 182, where asome- a tax unit, andwe cannotassumethat it is implied, sincekephalaiare other-
what different form ofentry is found, even though there is little doubt that wisealiento theinscription.InAstypalaeano. 180,thecountofhumancapita
it forms part of the same inscription. No totals of zugokephalai follow the contains many complex fractions. Owing to this, the consolidated capita
owners name or the entries of individual properties. The balance of the in theMagnesia inscription cannot beinterpreted; Jones's suggestion that the
fragment deserves interpretation. whole numbers are men while the fractions are animals must be rejected. 26
8) Magnesla ontheMeander:0. Kern, DieInschriftenvanMagnesiaamMeander The inscriptions therefore substantiate the argument about capita made
(Berlin, 1900), no. 122 (eleven fragments); Deleage, Capitation, pp. 194-5 earlier. There were schedulesfor converting personsandanimalsinto capita
(sample). The inscription is organized on the basis ofan alphabetical listing of assessment. These capita were equivalent to iuga, that is, they could be
ofplace names in the district and includes parts ofletters A, B, and B. The
model for each entry is as follows: "Such and such a place, according to 23 For example, e. 17, where no capita appear in a listing ofover ten iuga. OnJones s
inferencesregardingcapitain this inscription, below, n. 26.
24 Not a Christian priest. His entry was read differently byJones, "Records, " p. 57, and
20. ThereadingsofeventherevisededitionshouldbecomparedwiththoseofDeleage T. R. S. Broughton, Roman Asia Minor, in T. Frank, Economic Survey ofthe Roman Empire,
pp. 191-3- IV, 689.
21 Ibid., pp. 191-2. 25 Neither Deleage, Capitation, pp. 189-90, norJones, "Records, " p. 54, commented on
22 Ibid., p. 192, where the implication of the fractional human heads is not explored. this curious fact, which contrasts with the practice at Astypalea.
This column ofthe inscription was kindly verified for me in 1968 by Dr. Jochen Twele 26 Jones,"Records," pp. 52, 54f., 56f.JustasLot attemptedto arrive at the concrete
at the Institute for Advanced Study, Princeton; I am much obliged to him. The fractions realitieslyingbehindthecapitaofAutun (above,ch.N, n. 29),sodidJonesdrawdemo-
descendto 24ths,40ths,200ths, 50oths, 850ths.Dr.Twele'sreadingsagreewiththoseof graphicconclusionsfromtheMagnesiainscription.Neitherargumentcanbeentertained;
Deleage (p. 193) except at line 8. we lack the indispensable schedules.
 120 CapU t AND COLONATE
APPENDIX I: THE CENSUS INSCRIPTIONS 121

added on a ratio ofone to one, asshown by Astypalea no. 180; but the rates presupposirion of bureaucratic homogeneity and method. Delfage would
at which persons were evaluated as shares ofassessment are as unknown to have us believe that each inscription represents a step in a systematic process
us as the rates for animals. ofcadastration that was carried out in all the localities. This is highly doubt-
b) Were coloni registered on estates? The inscriptions abundantly illustrate ful, especially in view of the consignment to stone (why should they all
the registry of slaves, livestock, and inquilini - what we would call "cot- differ over which "step" to inscribe?). I would argue on the contrary that
tagers, " or tenants rent'ng shelter as well as land. In the Lesbos inscription, diversity andlocal options were the rule; the authorities managed asbestthey
tenantsarementionedbynameonly, butit remainsto beestablishedwhether could and were not guided by instructions from the capital. Lesbos no. 79,
their appearanceisin the guiseofelements ofassessment or (asI believe) of Astypalea no. 182, and Thera no. 343 illustrate that homogeneity was not
declarants ofthe property in question. Jones held that the capita ofthe Tralles attainedeven in the records ofone place.
inscription- those not specifiedto be slaves or animals- were coloni;"this e) Who made declarations? The inscriptions &om Magnesia and Tralles
can be neither affirmednor denied. On balance, it would seem that, at the feature all manner of distinguished men. The rest, excepting only that of
undetermined time when the inscriptions were carved, tenants were not Hypaepa, record substantial holdings, consisting of course of scattered
registered on estates unless they also leased the roof over their heads. 28 properties. Unless this is merely accidental, one might conclude that the
c) The rarity of qualitative distinctions of land. In the five instances where census did not aim to record all the land ofa district but only the land (and
land is expressed in iuga, we do not know whether qualitative distinctions heads) declared by men making declarations, the assumption being that
weremadeor not, sincethey wereanelement oftheconversioninto shares liability extended only to a limited segment ofthe population. Quite apart
ofassessment.In the others, however,distinctionsofthis sort occurnowhere from their utilitarian character, the census inscriptions may have been re-
else than in one ofthe seven Thera fragments. garded as a mark ofdistinction by those entered in them.
d) The process ofregistry. Deleage's valuable commentary ontheseinscrip-
tions is cast in such a way as to illustrate the system by which late Roman In closing, it may be appropriate to observe how odd it is thatSuch inscrip-
authorities compiled what he and many others have called "cadasters"; tions exist at all. Stone hardly seems a suitable material for registering data
certain inscriptions illustrate "primary cadasters, " others illustrate "second- that is bound to change from year to year. Livestock, especially in the form
ary cadasters. 29What should be questioned in this presentation is the tacit of sheep and pigs, is not a stable element of rural fortunes, and yet it is
recorded down to infinitesimal fractions of shares of assessment. How
27 "Records, " p. 57. permanent a record could these have been expected to be? One thing at
28 Cf.above,ch.m,n. 16andch.v, n. 14.A recentlypublishedpapyrusof'^39-40offers least is plain. The inscriptions that have been found come from a narrow
interesting new information on estatetenants leasing their accommodations andon the
ratherheavylabourservicestheyowedonthisaccount:P.Mich.620,inJohnC. Skelton, area within a single diocese; the localities in question are likely to have been
Papyrifromthe MichiganCollection, AmericanStudiesin Papyrology9 (Toronto, 1971). in close contact with one another. 80 It is as though the idea of carving the
29 Capitation,pp. 164, i6p, 181,194(thechaptersub-headings);theimpressionofsyste- census on stone had been a momentary regional fad, abandoned in one place
matic order iseven more rigorously given byKarayannopoulos, Finamwesen, pp. 43-53, (Chios) before it was even completed.
whosetables combineandorganizeelementsfrom differentinscriptions. Cerati, Came.
annon.p. 395,admittedDeleage'spresentationasa possibility,"maiscen'estpasextreme-
meat certain. On the inappropriateness ofthe term "cadaster," see my comment in
From Rom. Taxat.to Med. Seign.," p. 377n. 121,to whichI nowadda reminderthat 30 The provincesrepresentedareCaria,Lydia,andtheIslands,all in Asiana.But for the
theword"cadaster"itselfdoesnot antedatethelatemedievalperiod(seeA. Andreades, interveningsea,whichis betterenvisagedasa bridgethana barrier,thecircleencompas-
Deuxlivres recents surlesfinancesbyzantines, " Byzantinische Zeitschrift, 28 [1928], 304). sing these localities would be very small indeed.
Yet oneencounters "cadaster"in everystudy dealingwiththissubject, asin therecent
excellent work ofN.G. Svoronos, "Recherchessur 1c cadastrebyzantin." If historians
stopped invoking anachronistic "cadasters" and used instead some appropriate terms
describing, not whatlay in thefuture, but the documents theywere looking at, thenwe
might be a step nearer to a general history of these difficult administrative records.
Besides, hardly anyone nowadays knows what even a modem "cadaster" is.
 APPENDIX IF: SUPPLEMENTARY NOTES 123

bringgreaterequityto localassessmentanddistributionofthetaxburden.Theappro-
priate setting for changesin the total waseither imperial fiat (scalingthe sum down-
wards or upwards) or appeals by local authorities (municipal or provincial) to the
imperial government.
Appendix II n. 61 It is illusory to suppose that municipal aristocracies would, over a period of
centuries, have sustained enthusiasm for service to their patriae if that service had not
entailed opportunities for private advantage, whether from incentives offered by the
emperors (such as Roman citizenship) or from profits, legitimate or otherwise. The
abuses that Salvian accuses curials ofmaking oftheir position (Degiibern. Dei 5. 17-18)
were not a fifth-century novelty. But private profit doesnot at all times exclude com-
munal advantage.

SUPPLEMENTARY NOTES n. 72 To Abbott andjotmson, the procedure ofnominatio laid down in Lex Malacitana
5i (A. D. 82-4) bore out that the decay of the municipalities had "already" begun
CHAPTER I (Munic. Admin., p. 112). A debate has recently taken place over the question whether
Pliny Epist. 10. 113 should beretained asthe terminus a quo for reluctance on the part of
n. 19 Theevidencethatlandandheadsweretaxedisplain(e.g.,CassiusDio62. 3.2-3; citizens to undertake activities on behalf of their cities: G. Bowersock, Augustus and
Digest 50. 15. 8. 7, which is regarded asthe lows classicus). The questionable point ishow the Creek World (Oxford, 1965), pp. 148-9; A. N. Sherwin-White, The Letters of
suchevidenceshouldbe interpreted in reconstructing the form ofimperialrevenue- Pliny (Oxford, 1966), p. 724; C. P.Jones in Phoenix, 22 (1968), 137-8; F.A. Lepper in
raisingfrom theprovinces.Thestatementthat"DieHauptabgabederProvinzenwar Gnomon, 42 (1970), 571. Such a pattern of interpretation is highly questionable. To
dieGrundsteuer,tributumagri"(Schwahn,"Tributum," col. 10)isverydifferentfrom interfere with an institution means, no doubt, that it is somehow defective; but what
thealternativestatementthattribuhimsoliwastheprincipaltaxleviedlocally in order institution is not? The point of intervention is to improve something capable ofim-
to covertheamountsowedbytheprovincesto theimperialgovernment.Inthelatter provement - an indication, therefore, of the fundamental health and viability of the
case, the imperial government's primary interest waswith total payments by localities; institution being tinkered with. No logic decrees that municipalities must have been
in the former, its solicitude extendedto the tax itself. more effective before rather than after the emperors instituted regulatory measures;
one would first have to establish that they had been. In this perspective, the ominous
n. 26 Hence the report ofTacitus Ann. 2.42. 5 that Syria andjudaea, exhausted by moment for municipalities came no earlier than when the emperors began to issue
onera,i.e.,taxesofmanykinds,pleadedwiththeemperorfora loweringof"tribute," regulations tending to undermine municipal cohesion (e. g., CTh 12. 1.23-7, 6.22. 2).
i.e., ofthetotalexpectedfromthem.Ifthereductionweregranted,it couldbeimple- When this happens, we acquire an unambiguous basis for talking about municipal
mented in a variety ofways: cancellation ofsome individual tax, lowering ofthe rate decline.
ofseveral, etc. The same complexities in implementation are implied, though regret-
tably never spelled out, in the caseswherewehearofreductions intotals (below, n. 30). CHAPTER II
Was implementation left to local authorities?
n. 8 Jones, LRE,p. So, took Constantineto havemerelyexemptedthe clergy from
n. 44 Schwahn, Tributum," col. 11, listed the many provinces (Spain, the Gauls, curial duties; hence, tax exemption came only in the next reign (p. 118). Gaudemet,
Macedonia,Illyricum, Achaeia,Syria,Kyrene, andAfrica) from which"wurdevon "Legis. relig., " pp. 28-32, and Dupont, "Privileges, pp. 733-5, realised that munera
Anfang an eine feste Steuer, stipendium oder t[ributum], nach einem Pauschalan- have broader implications, but they insisted on contradicting the evidence of the
schlag [= lump sum] gefordet ..." (Abgabe is preferable to Steuer in this context.) Syro-Roman Lawbook; IIn est question que demunera ou de liturgies. Lesimmuni-
Thiswasdone,hestated, "wonochkeineallgemeineLandesvermessungvorlag," the tes d'impot n'apparaitront que sous les fils de Constantin" (Gaudemet). These con-
implicationbeingthat, oncea taxsurveyhadtakenplace,thetotal wouldbealtered. clusions deserve reconsideration. First, many laws prove that munus and especially
But there is no reason to suppose that a census resulted in changmg the total; even functio are terms that embracetaxes as well asliturgies; cf. Grelle, Stipendium,p. 44.
today, communitiesengagingin extensivereassessmentfor purposesofproperty tax A restrictive reading of munus isjustified in some contexts, but the term cannot be
calm the fearsoftaxpayersby assuringthem that the object ofthe exerciseis not an taken as restrictive in principle. The most sweeping form ofexemption from public
increaseoftotalproceeds.Thepointofa censusaftera lump sumhadbeensetwasto burdens wasnecessa. rily expressed asfreedom from munera, immunity - a point that
 124 CapU t AND COLONATB APPENDIX II: SUPPLEMENTARY NOTES 12$

is most compendiously illustrated by CTh 13.5. 5 (3295); "Navicularios omnes per name, regardless ofwhether he was a resident or not), it being understood that, in the
orbem terrarum per omne aevum ab omnibus oneribus et muneribus, cuiuscumque locality where one resided, one's patrimony obviously stood pledge for one's per-
fuerint loci vel dignitatis, securosvacuos immunesqueessepraecipimus ... ut a con- formance ofmunera. Sincea rule ofthis kind would have limited the advantagesof
lationibus et omnibus oblationibus liberati integris patrimoniis navicularium munus absentee ownership within a setting where the performance ofmunera was central to
exerceant. " If it were not for the final clause, we might well infer, as is done with taxation, it would have constituted a very wise safeguard to the health ofcommunal
regard to the clergy, that onera and munera excluded taxes. (Are there any texts where finances. Moreover, if my suggestion is correct, then the vacatio enjoyed by veterans
muneraarecontrastedto "taxes"?)Constantine'splainintentwithregardto theclergy underDiocletianwouldhavesufficedto makethemfully immunewheretheylived,
was that they should enjoy complete freedom from care in exercising the clerical but not wherethey ownedproperty as absentees;andit wouldhaverepresentedno
munus. appreciablediminutionofprivilege by comparisonwith thatenjoyed byveterans in
A secondreasonforreconsiderationisthattheauthorscited (andothers)readthe thetimeofAugustus.Ininterpretingprivileges,I amtemptedtoproposetheguideline
post-Constanrinianlawsonclericalimmunityasextensionsofprivilege.Thepreferable that privileges in general are stable but the setting within which they are exercised is
alternativeisto readthemasa seriesofclarificationsdesignedto rendermore specific not; thus what should be looked for in the texts are changes in the setting, or admini-
the original,simplyphrasedlawofConstantine.Tax-collectorswereunderstandably strativecontext, ratherthanin theprivilegeitself. (Seefurtherbelow,ch.iv, n. 42*.)
inclined to encroach upon clerical immunity, by alleging, for instance, that Constan-
tines lawdidnotextendtonovaecollationes(CTh 16.2. 8); theyhadto becontinually n. 23 The laws ofAurelian and Constmtine might be taken to imply that there was a
held at bay. The breadthand comprehensiveness ofConstantine's initial exemption land tax ; so might the law declaring veterans' land immune (above, n. 12), and
ofthe clergy is best inferred from the radical restriction he himselfset upon it in 329 others like it. When land is the predominant asset, one easily gets the impression that it
(s), by denying entrance into the clergy to those best qualified to bear the saecuh is taxed. The problem is how to define "land tax. " For example, if men are taxed
necessitates {CTh 16. 2.6 and 3); see p. 29. according to their declared assets, andifthese assetsinclude land, should weunderstand
that the land they own is subject to "land tax"? Such a reading would arbitrarily
n. 10 T.C. Skeat(PapyrifromPanopoUs,p. xxx) saidthatthetwo papyri"donot add isolatelandfrom otherassets.Inthecaseofa veteranwhohadimmunityforhisland,
greatly to our knowledge of the level of taxation or the mechanism of collection." shouldweinterpretthisasmeaningthatanexistingtaxonthislandceasedto bepaid,
Another view might be that both papyri take us to the heart of the Diocletianic tax or that the veteran was entitled not to declare hisland to the census? To my mind, the
process and enable us to see it not as a hygienic levy of fixed amounts from fixed distinction to focus upon is that between land as security for a personal debt and land
quantities ofland, butasanalmostdailyexactionofvariable commodities, money, and that pays (below, ch. m, n. 9). In the first case (the one I regard as applying to the
services by imperial officials from the political groupings in which the taxpayers lived. lawsjust mentioned), the tax weighsupon the land only in default ofpayment by
TheimpendingimperialvisitthatovershadowsP. Beatty Panop.I showsthemachin- the registered taxpayer; in the second, taxes are a first charge upon the land, an obliga-
cry when operating under the stram ofextraordinary exaction, but the more routine tion necessarily assumed together with possession.
itemsinterspersedwithintheletter booksufficeto provethecentralityofnomination
and performance even in the ordinary tax process. n. 24 In commenting upon the last sentence of this law - "Opulentos enim saeculi
subire necessitates oportet, pauperes ecclesiarum divitiis sustentari" - J. Gaudemet,
n. 14 Whenthisbookwasinpress,I cameacrosstheimportantdiscussionofmilitary "Constantin et les curies municipales, " Iwa, 2 (1951), 56 n. 55, said: "C'est aussi la
privileges by D. van Berchem, L'armee de Diocletien et la reforme constantinienne, premiere mention de la richesse des eglises"; cf. A. Piganiol, L'Empire chretien (Paris,
Institutfransaisd'archeologiedeBeyrouth,Bibliothequearcheologiqueet historique 1946), p. 368. It is probably anachronisticto understandeccksiwumdivitiain the sense
56(Paris,1952),pp.75-8.InvanBerchem'sview,theimmunityofveteranshadbythe ofchurch endowment. Although great churches, notably that ofRome, were richly
third century been confined to munera personalia, leaving them subject to munera endowed at an early date, it is more significant for the general financial condition of
patrimoniorum; as a. result, veterans' privileges were stated under Diocletian to consist the churchthat episcopalandmona.stic endowmentswere still regardedasa novelty
in honorum et munerum personalium vacatio" (CJ 10.55. 3; cf. 7.64.9, 10.55.2; in early fifth-century Africa: Vita s. Melaniae 20, ed. D. Gorce, Sources chretiennes 90
quoted by van Berchem, p. 77). VanBerchem's clearpresentation suggests a question (Paris, 1962), pp. 168-71. The legislator is better interpreted as referring not to con-
(which he did not ask) that may seem paradoxical, namely whether one bore munerd solidated wealth but to the capacity of Christian communities to generate charitable
patrimoniorum in the place where one was subject to munera personalia - in other words, contributions.
whetherthelattercategorymightnothaveincludedtheformerinanygiven locality.
Ifso, then a veteran's, or anyoneelse's, liability to munerapatrimoniommwouldhave n. 26 The essentiallink between"liturgies"andtaxationhasbeenconsistentlyover-
been limited to localities where he was an absentee landowner (i.e., where, in the looked in modern accounts. Typically, liturgies are dismissed in a footnote in Ardant,
language ofDig. 50. 15. 4, a property owner was held vinmlis censualibus in his own Hist. de I'impot, I, 66 n. I. Another result is the inference that any stress upon munera
 126 CapU t AND COtONATE
APPENDIX II: SUPPLEMENTARYNOTES 127
is a novelty documenting financial difficulties. Oertel (as above, ch. I, n. 72), who Mga as consisting of Crundstiicke of varying size but equal in Einkommenswert (cf.
invoked hated 'liturgization, ' " wasa characteristic spokesman for thistendency.A Stevensin Camb. Ec. Hist.,i2, 113). Somecommentatorsreifiedtheiugumto such an
greatmerit ofGrelle's exposition (asabove, ch.I, n. 23) isthatit corrects thisdefective extentastosupposethatit representedtheamountoflandnecessaryfora single-family
perspective.
farm; accordingly, F. Lot disputed the figures of the Lawbook on the grounds that
more land was needed to sustain a family: "De 1'etendue et dela valeur du caput fiscal
CHAPTER III
au Bas-Empire, Revue historique de droitfranyis et etranger, 4th ser., 4 (1925), 19-23.
The actual process was that, by the application of a schedule, such as that of the
n. 2 J.-P. Callu, Lapolitique monetam desemperews romams de 2380311, Bibliotheque Lawbook, landed property was converted into uniform shares of assessment; these
des Ecoles frans-aises d'Athenes et de Rome 214 (Paris, 1969), pp. 290-300, 319-23, shares were iuga, andthey existed nowhere else than in the tax records. The texts often
though interesting in details, gave the general impression that taxation was a distinguish real lands, possessiones, from the assessment, professio or iuga, that they
strictly imperial matter. R. I. Frank, "Ammianus on Roman Taxation, " American represent. See my remarks in From Rom. Taxat. to Med. Seign., " pp. 172, 174, and
JournalofPhilology, 93 (1972), 70-2, usefully indicated thedifferences between second- add CTh 5. 11. 8 (365) to the evidence cited there. One may legitimately attempt to
andthird-centurypracticesin levying theannona;seebelow,n. 8.Thecharacterofthe discover the scale offortune that underlay the iugum, but while this information may
indictiopriorto beingreformed (i.e.,itsbeinga suddenrequisition) waswelldesigned be valuable, it adds little to our understanding ofthe assessment scheme.
to enhance the power of local authoritites.
Asindicatedabove(ch.II,n. 10),P. BeattyPanop.1-2areofexceptionalinterest n. 8 Jones, LRE, pp. 1191-2 n. 110, argued against the contention (e.g., Piganiol,
in illustratinghowimperialandlocalofEcialsinteractedin thecollectionanddistribu- Emp. chret., p. 337) that "the praetorianprefecture merely fixedtotals for provinces
tion ofthe annona . An astonishingly large role is played in these papyri by money or cities and that it was left to the provincial or local authorities to share out the total
alongside commodities; P. 2, lines 229-44, illustrates among other things that the between the taxpayers ; he cited passages from Themistius {LRE, p. 1189n. 101) and
practice of adaeratio, and the schedules that made it possible, existed from the very Justinian {Nov. 128. 1) stating that the rate per share ofassessment was fixed in detail.
beginningofthelateimperialtaxsystem(cf. below,ch.v, n. 54*). It is true that, if the prefecture had records oftotals ofshares, it could match them
against the total expenditure estimates and come up with rates per share. But these
n. 4 CJ 10. 16. 3 (249), sinceindictiones apply to property, notpersonae, the provincial were shares ofassessment, not persons or even land; the prefecture's establishing rates
governor will protect the petitioner from beingchargedbeyondthepro rataofhis persharewassimply the meansfor arriving at totals. Theproblem for theprovinces
property. C&ati,Came.annon.,pp. 516-17,in suggestingthatthisrescriptappliesto and especially the municipalities consisted in finding the taxpayers corresponding to
someone owning property in several provinces, overlooked the fact that conlatio the shares; this was likely to be an active process, since the census was almost always
wasnot provincial but municipal. The rescript appears to portray a situation in which out of date. The fact that the prefecture established rates per share did not turn the
the petitioner is being asked, in view ofhispersona, to contribute more thanhisshare, local authorities into mechanical executors of orders; c£ the sharp complaints of the
ontheprinciplethatwewouldcallnoblesseoblige.Theemperor,whocouldnottariff procurator in P. Beatty Panop. 2, lines 229-44, who refers precisely to how little
generosity, gives the inevitable reply: contribution to mdictiones is proportional to difference it appeared to have made that maximum rates per commodity had been
assessment only. laid down by "divine decree, i.e., the year's delegatio (and not the edict of 297, as
Skeat believed).
n. 6 A schedule similar in form to that ofthe Lawbook is found in the Byzantine
compilation ofunknown date called the Ceometrica ofHero ofAlexandria, ed. Wil- n. 10 A crucial issuein the interpretation ofthis text, which a dictionary cannot settle,
helm Schmidt andJ.L Heilberg, Heronis Alexandriae opera, iv (Leipzig, 1912), 414. is whether one should adopt the reading ofBoak "amount levied" (on each aroura,
Here, various types of earth - black, red, hard, sandy, saltpetre-containing - are on each head), or that ofDeleage, Capitation, p. 43, taux d imposition, " in English,
classed by iugerain relation to a unit ofassessment most often. calledgeikon, but also rate. Sincethe object ofDiocletians law was to eliminate inequalities in assessment,
kephale. The equivalence ofthe two units isclear; for example, "looiugera ofnewly Deleages reading is almost certainly the correct one. The nature of taxation at the
planted vineland amount to i geikon .. 100 iugera of hay meadow are I kephale. time- namely, thevariabilityandflexibilityoflevies,illustrated, forexample,by the
Owing to the uncertainties of date and provenance surrounding this text, and its Panopolispapyri- excludestheideaofa fixedcontributionperarouraandperperson.
remoteness indetailsfrom theotherevidenceavailabletous,thereseemsto benoway Boaks commentaryshowsa persistentinabilityto differentiateassessmentfrom actual
ofutilizingits testimony. taxation.Thus,heconcluded (p. 25), "Theevidencefor botha landtaxanda capita-
tion tax in the edict is incontrovertible"; so also Karayannopoulos, "Die Theorie
n 7 A. Piganiol, Histoire de Rome (Paris, 1939), p. 446, spoke of "unites foncieres Piganiols, " p. 342. This inference hinges on reading "amount levied. " If "rate" is
d egalevaleur," whileKarayannopoulos,"DieTheoriePiganiols," p. 325,mentioned actually meant, then the edict proves only that persons as well asland entered into the
 128 CapU t AND COtONATB APPENDIX II: SUPPLEMENTARY NOTES 129

computation ofassessment.The personalcomponentofthe edict is discussedbelow, by Cerati) asmeaning that census assuch applied to fields andnot to persons: precisely
ch. iv, n. 22. because the munera being mentioned in this part ofthe law (parag. 3) pertain to patri-
It was assumed by A.C. Johnson, "Lucius Domitius Domitianus Augustus, monies, no one may excuse himselfon grounds ofpersonal or ecclesiastical privilege
ClassicalPhllology,45 (1950), 17,andBoakandYoutie,Arch.ofAw. Isid.,p. 24,that from complying with the regulations being prescribed.
the evil practice"referredto wasthattherichshiftedthe burdenoftaxationupon
the poor. This is an instinctive modem inference (below, ch. v, n. 59), but not one n. 23 M. Pallasse, Orient et Ocddent a propos du colonat romam du Bas-Empire (Lyons,
that should be lightly drawn, since the text may be otherwise interpreted. IfA and B 1950), pp. 28-35, sought to establish a fundamental regional distinction in the status of
were ofequal fortune, small or large, and if, owing to confusions in assessment, B coloni by contrasting the western term capitatioplebeia (never mentioned after 368)
paidmore thanA, thentheconditionsstatedby the prefectareadequatelyfulfilled. with the easternterm capitatiohumana (first mentioned in 393). Onthe latter, below,
On this reading, the edict testifies only to a problem in tax assessment, and not to ch. iv.
social abuses.
n. 36 Deleage, Capitation, p. 219, construed iugorum capitationibus as "les evaluations
n. ii Thoughit isoftensaidthattheiugumwasnotintroducedintoEgypt,thesilence de leurs unites-jougs en vue de la capitation. " I would translate, "the unsecured and
ofthe papyri doesnot prove this, sincethewgwndidnot supersede units ofland mea- secured declarations of shares of assessment. " See also CTh 7. 13. 7. 1 (375): "ut it
sureeitherinEgyptoranywhereelse; inthecaseofarableland,justasthewgumwas tantum a consortibus segregentur, quorum iugatio ita magna. est, ut accipere non
expressedmplethra oriugerain Syria,sowasit statedinarourasinEgypt. It ispossible, possit adiunctum [scil. in exhibendo tirone] ... Sin vero aurum fuerit pro tironibus
however, that Africa and southern Italy were differently treated, for we hear ofthe inferendum, unumquemque pro modo capitationis suae debitum redhibere oportet.
centuria in the former, CTh 11. 28. 13 (422), and the millena in the latter, NValent The payment ofgold on the basis of. capitatio was unquestionably intended to extend
5. 1.4 (440). usedlike wgaasshares ofassessment. Owing to its late date, this evidence to thelargelandownerswhosehigatioisearliermentioned.Segre,"Studies," p. 117£,
may testify only to thedisplacementoftheiugumbycomparablylargelocalunitsof welcomed capitationes wgorwn on the basis ofthis text.
assessment. See also above. AppendixI, nn. 1-3. Deleage, Capitation, pp. 115-18,
reached interesting results in his computation of how many arouras made up a CHAPTER IV
wgum.
n. 4 It has received many readings, among them the liberties ofFustel de Coulanges,
n. 14 What I intimated there, p. 175n. 54, about the disappearance ofthe wgum in L'alleu et Ie domaine rural, 4th ed. (Paris, 1927), p. 83, Si Ie proprietaire a neglige de
Byzantiumis borne out by FranzDolger, Beitrcige zw Geschichte der byzantmischen declarer un fermier, il est responsable de 1'impot pour cette parcelle. " This pre-
Finanzverwaltung, Byzantinisches Archiv 9 (Leipzig, 1927), p. 54. Nevertheless, the supposes thatthe object in question is a peasant tenure, andnot the landlord's complete
subject deserves further study, especially in the light of the Byzantine sources and holdings, as Ulpian plainly indicates. Clausing, Colonate, p. 101,translated, Ifanyone
secondaryliterature. Forexample, thereferencestoannonocapitaintheLiberpontificalis, has not declared his inquilinus or colonus, he will be held accountable by census laws."
VitaJohannis v 2, and Vita Cononis 3 (ed. L. Duchesne, i, 366, 368-9), were inter- This ispossible asto words, but isifanything more inaccurate than Fustel, who rightly
preted by Ernst Stein, "Zur Geschichte der byzantinischen Finanzverwaltung," saw that the pouit of the rule was to fix responsibility for public charges; unless an
Vierteljahrschriftfur Sozial-undWirtschaftsgeschichte, 21 (1928-9), 159-60,andG. Ostro- agentwasdeclared,the ownerborethe responsibility.
gorsky, Das Steuersystemin byz. Altertum und Mittelalter, " Byzantion,6 (1931), On the meaningofinquilmus (much discussedin the literature on the colonate), I
237-8,asillustrating thecontinuance tothe68osandbeyondofthefourth-centurytax agree with A. H. M.Jones, "The Roman Colonate, " PastandPresent, no. 13 (1958), 3-4
system in its original form; cf. Karayannopoulos, "DieTheorie Piganiols, " pp. 334-5. (cf. LRE, pp. 796, 799), in adopting the standard definition, for which see Dig.
I believe that this conclusion is mistaken but would like to reserve discussion for a l9-2. l9. 4~5; 43-32-!; CJ 6. 31. 1 (214). It is also the definition of Augustine Enarr. in
future publication. Psalm. 118. 8. 19. 1, "Inquilini non habentes propriam domum, habitant in aliena.
(Corp. Christ., Set. Lat., XL, 1684). Cf. CTh 9.42. 7 (369), quot sint casariivel coloni,"
n. 15 C&ati, Carac. Annon., pp. 319-28, surveyed five laws that, in his view, force distinguishing tenants into categories living on and offthe estate.
one to recognize the existence of a distinct head tax, namely, CTh 13. 10.2, 13.4.4,
cj ii.52.i, cr/i 12.4. 1 withNT/; 22.2. 12, andNValent10.All butthe lastoftheseare n. 9 The activities of the censitor Julius Sabinus in Upper Egypt (Boak and Youtie,
dealt with elsewhere. The key phrase in the law ofValentinian in reads, "census, qui p. 30) do notjustify the inference ofan Empire-wide census. The miscellaneous evi-
non personarum est, sed agrorum. " Cerati concluded that, since the assessment of dence for censusesunder Diocletian (Deleage, Capitation, pp. 44-66, 152-7) merely
fields was distinguished from the assessment of persons, there must have been a illustratesthatcensuseshadto beoccasionallyrenewed.Infact,theevidencefor Syria
separateassessmentofpersons(p. 327).I preferto readthephrase(in a wayrejected and Arabia documents the delimitation of city territories, an operation that has no
 130 CapU t AND COLONATB APPENDIX II: SUPPLEMENTARY NOTES 131

demonstratedconnection with an internal registry oftaxableassessment. C£ Grelle, personal declarations was the basis for the levy of a tax apart from a roster ofland;
Stipendium, pp. 26-7. M. Adinolfi, "Una iscrizione greca. inedita e la capitatio di the only evidence for such a levy is the four receipts, and they, aswe have seen, relate
Diocleziano, Oriens Antiquus, 4 (1965), 71-5, claimed that the Syrian inscription he to a city. A puzzling situation, but not unique: the Tralles inscription (seeAppendix i)
edited "determines theformula census of a given village into fivetaxablelandunits, lists and counts heads (even of animals), but its total consists ofwga only. I would
into five iuga (p. 75). The inscription says no more than that five boundary stones suggest that, in line with the bulk ofthe papyri, it should be concluded that Egyptian
(oroi) were set up for the prosperity and victory of the emperors and Caesars. To assessment was based on land alone. The schedule ofcapitawas introduced into Egypt,
jump from this to iuga, which are nowhere else related to boundary stones, seems but owing to the millennial tradition ofland survey and registry it served no useful
arbitrary. R. I. Frank, "Ammianus on Roman Taxation, " pp. 72-3, rightly indicated purpose there and was never adopted asan element ofassessment. Callu, Poht. monet.,
that censuses were carried out locally but added, without apparent evidence, that these pp. 290 and 321 n. 6, asserted on the basis oftwo ofthe receipts for city dwellers that
records were copied in full and sent onward to the governor and Prefect; only then, Diocletian continued the tributum capitis throughout the Empi reand that it was
allegedly, would the contents have been condensed (presumably into tax units). somehow integrated within thenew scheme. This wasalsothe viewofFaure, "Etude
de la capitation (below, n. 30*), pp. 110-16, and ofKarayannopoulos, "Die Theorie
n. ii The early fourth-century Egyptianevidence drawsour attention to a money Pigaaiols, pp. 342-4, who alleged as fact that a personal tax distinct from land tax
tax vulgarly known as the delegatio and distinguished from the annona (P. Cair. traversed the whole early Byzantine period. Stein, Zur Gesch. der byz. Finanzw.,"
Isidor. 42, 102, i03; and cf. the money taxes in P. Oxy. 1905). This refers presumably p. 160, stressed continuity from thetributum capitis ofthe Prindpate. I havepointed out
to the items other than grainandfodderthat the taxpayerswerecalledonto supply (ch. i) that tributumcapitisshould be regardednot asan imperial tax but rather as a
by the annual imperial dekgatio, whichwere (or could be) commuted to money in form of direct assessment that localities could impose with imperial authorization in
Egypt. A money tax ofthis kind still fell within the general rubric otannona and did order to meet their tributary totals. If this is so, then tribiitum capitis drops out ofthe
not constitute a distinct "capitation." category ofanempire-wide institution, and oneneedno longer expect andlook for its
continuity into the later Empire.
n. 2i For instances ofangariae, P. Beatty Panop. 2, lines 153-5 (suppliers invited to
reclaim carts and cattle sent to quarries), also lines 47-8 (villagers having to tow state n. 26 If this were a standard caput, the surprising implication would be that Sidonius
boats upstream). For confirmation of the view that the "rural commoners" were bore the very small total liability offifteen capita (if he sought a twenty per cent tax
takinganupwardstepby registeringfor theannona,seeDig. 50.5. 1.2 (Ulpian): "Qui cut) or 30capita(supposingtenpercent). Thealternativeisthatcapitarefersto "heads"
in fraudem ordinis in honoribus gerendis, cum inter eos ad primes honores creari oftaxation, what the laws call tituli. P. Ital. i, for example, shows that a high ofEcial
possint qui in civitate munerabantur, evitandorum maiorum onerum gratia ad colonos (ca. 445) caused a rite/tu oftaxation for one of his Sicilian estates, worth 75 solidi, to
praediorum se transtulerunt, ut minoribus subiciantur, hanc excusationem sibi non beturned into anitem ofhisrent. Sidonius may have sought a favour ofthe same sort.
paraverunt. Beforehim. St. Basil (Epist. 285) hadalreadycalledtaxationa "hydrawithmultiple
heads, " as illustrated by P. Oxy. 1905, above, ch. m, a. 8. Lot, "Etendue et valeur,"
n. 22 The law 0^377 (CTh 7 6. 3) proving the equivalence ofcapt ifand wgum indicates p. 32, rightly pointed out that, poetically speaking, the interpretation of capita as
thatthere wereno capitain Egypt;andDeleage,Capitation,pp. 99-101,showedthat tituli is preferable. This may explain how, eventually, Gregory of Tours would use
the papyrus that hadformerly been thought to document the existence of capi ta did the name capitularius for the "librum in quo census huic populo est inflictum" (Hist.
nothing of the kind (the "heads" in question are units receiving rations). (See also 9. 30). For another instance ofcapita used generally (Illyricum, 412), see below, n. 72.
below,n. 38*.) Althoughalmost all Egyptiantaxationis relatedto landexpressedin An even later, and again Gallican, reference to capita hascome to my attention in
arouras,Deleage,pp. 44-8, elaboratedon the prefect'sreferenceto "eachheadofthe the Testamentum s. Caesarii Arelatensis, ed. G. Morin, S. Caesarii opera omnia, H
rural population" bydiscussingtwo personaldeclarationsby villagers (dated309and (Maredsous, 1942), p. 289: pietas divina concessit, ut per meam humilitatem im-
310) and four receipts for payments ofpoll tax by city dwellers (dated from 301to munitas ecclesiae in tot capitibus daretur"; cf. p. 287, "deus misericors per parvitatem
314). The same papyri have also had a large place in other discussions ofDiocletian's meam etiam immunitatem tributorum ... ex maxima parte concesserit. (The Vita
reforms.I doubtthattheyarehelpful. Sincethereceiptsareforcitydwellers,theyare Caesarii 1.20, ed. Morui, p. 304, specifies that the Visigothic king Alaric n "dati
likely to be irrelevant. The prefect says nothing about city dwellers, while, on the firmitate praecepti ecclesiam in perpetuum tributis fecit immunem. ") That capitahere
other hand, there could betaxes that did not form part ofthe new scheme, such as the should, as in Sidonius, refer to tituli oftaxation fits the context. But it is also possible
olderheadtaxmentionedbyDeleage(cf. BoakandYoutie, Arch.of AM. Isid.,p. 25). that Caesarius means caput in the sense of a "share of (landed) assessment"; thus, tot
As for the two declarations (one of them is P. Calr. Isidor. 8), they are simply state- capita, so many shares ofassessment, were made immune as to liberate maximapars
ments of liability: so-and-so declares himself and those living with him as "taxable" ofthe churchs property from taxation.
or "exempt," andsuppliestheirnamesandages. We cannotinfer that the roster of Cerati, Carac. annon., pp. 339-40, read capitain CTh 11. 16. 4 (328) asthough it were
 132 CapU t AND COIONATB APPENDIXII: SUPPLEMENTARYNOTES 133
a taxterm.Thetextshouldactuallyberead,"scriberedebebunt...persingulacapita," ratepersinglekephale,thenthe totalperkephale,thentheamountowedfor oneand
i.e.,thegovernor'sorderregardingextraordinariaisto bewritten downunderdistinct one-sixthkephale,i.e.,theamountreceiptedbythecollector.TheletterofConstantine
chapterheadings," whosesubstancethelawspecifies.Thetextthereforefallsoutside showsthattheabstractcaput,"unsecuredshareofassessment," equivalenttotheiugum,
the discussion. couldbeappliedearlierthan359inanEgyptiansetting.Nomorewasneededto make
this true than to establish an ofEcial equivalence: x-many arouras (possibly distin-
n. 29 F. Lot, "Etendue et valeur, " and "tejugum. Iemanse et les exploitations agri- guishing categories offertility) = i caput(unsecured) = i iugum (secured). Theearlier
colesdela France moderne, " Melanges d'histoire offertsa Henri Pirenne (Brussels, 1926), papyri show that, even ifthis equivalence wasestablished somewhere, the administra-
I, 307-26, worked hard at translating these 32, 000 or 35, 000 capi ta into figures that tion in Egypt didnot useit but continued to announce taxrates per aroura. The text
wouldtestifyto demography,to the typicalsizeofpeasantholdings,andevento the of359documentsanadministrativechangeconsistinginannouncingratesinkephale,
extentofthemedievalmansus.Regrettably,nothingofthesortcanbedone.Theseare thusmakingit necessaryto convertarouraintokephale,byapplyingtheschedulethat
neither lands nor people but abstract shares ofassessment, and we lack the vital con- probablyhadbeendrawnup long beforebut, asit seems,neverused.This (rather
version schedules that might permit useful statistical speculations. Cf. GofFart, "From otiose)changemayhavebeenrelatedtothespiritofreformdiscernibleinsuchlawsas
Rom. Taxat.to Med. Seign.," pp. 167-75. CTh 11. 16. 7, 8 (356).

n. 30 I have seen E. Faure, "Etude de la capitation de Diocletien d'apres Ie pane- n.42 Theindispensable commentary onthemilitary implications oftheselawsisby
gyriquevm, Varw:Etudesdedroitromam,rv (Insritutdedroitromainde1'Universite van Berchem, Armee deDiocletien, pp. 78-86. Since I sawthis book only when mine
de Paris, 20; Paris, 1961), 1-153, whose principal conclusion is that the capita in wasinpress I takethisoccasiontocontinuetheremarksbeguninch.n,n.14*, asthey
questionhererepresentlandexclusively.Faure'sargumentarionissoinvolvedthatit is bear upon the Brigetio Table and Constantine's law of 325, which van Berchem
impossibleto dealsystematicallywithpointsofagreementanddisagreementbetween usefullyalignedwiththeearliermilitaryprivileges.VanBerchem'sinterpretationof
hisviewsandmine, exceptat anexcessivecostin time andspace(cf. Cerati, Carac. capita is borrowed from Deleage and others, who identified them with shares of
annon., pp. 509-39). The same holds true for his article "ItaUa annonaria. Notes sur assessmentequivalent to a iugim (below, nn. 44-5*). In my view, the two laws -
la fiscalite du Bas-Empire et son application dans les differentes regions de 1'Italie," military details aside- document the adaptation ofan old privilege to a newsetting:
Revueintemationaledesdroitsde I'antiquite,3rd ser., 11 (1964), 149-231. what Diocletian had expressed in terms of honores and munera personalia was now
phrasedinthenovelvocabularyotcapita,"headings"inthecensusofa locality.The
n. 38 Constantinestates. Ifeach[ofyourfollowers]isburdenedwiththeadditionof law of311 may therefore offer a noteworthy confirmation ofthe ideaI put forward
ten 'heads'andthetaxesonthem,hewill atoncesweatviolentlyunlesshetakesthe that the moment oftax reform, " or at least ofadministrative change, occurred in
quickest road to the church ofsalvation ... " (the reading kephalaion, Athanasius, must relation to Galeriuss census of 306 rather than in Diocletian's reign. The law also
bepreferredoverkephalon,Gelasius,butthereferenceto capitaseemsunmistakable); suggests that the caputas heading"- active citizen for purposes oftaxation - was
a few lines later, he repeatsthe threat in the form that Arius's followers have made thesuccessoroftheolderconceptofa citizenwhoperformedmunerapersonalia.
themselvesliabletothecuriaanditsliturgies.I amobligedto mycolleague.Professor
T. D. Bames of University College, University of Toronto, for assistance with this n. 67 C&ati, Carac.annon.,p. 319,citedthistext, coupledwithits predecessor, CTh
passage. For a commentary on the transmission, date, and context of the letter, see 12.4. 1 (428),asoneoffiveitemsprovingtheexistenceofa personaltax.Inmyview
Eduard Schwartz, Gesammelte Schriften, m (Berlin, 1959), 73-85, 239-46. Domes the two laws illustrate the phenomenon to be mentioned shortly, namely the dis-
spoke o£Kopfsteuer and Schwartz (p. 243) more generally of "exorbitanten Steuer- appearance ofthe caputasa shareofassessment. The lawof428imposed anassessment
zuschlagen. The threat clearly refers to sharesof assessmentthat werenecessarily pro singulisiugiset capitibus (as Cerati pointed out, it is the only lawwehavewhere
unsecuredandprobablyequivalenttoa wgumeach:thetaxesarisingfromtheseshares these terms arejoined by e(). The legislator thought his sense was clear; heunderstood
were to be crushing.With regardto the secondclause, Schwartzinferredfrom the iuga et capita as shares of assessment, by now wholly landed but still, for historical
referenceto curialliturgiesthatthoseofthefollowerswhowereclergymenwereto reasons, bearing two names. In 443, clarification was needed; the assessment was
havetheirimmunitywithdrawn.Whilethisispossible,theinterestingalternativeis imposed only on iugationes, not on humans, animals, or other moveables. What we
thata liabilityoftencapita(i.e.,capitavelwga)orthereaboutwasthequalificationfor mayinferto havehappenedin theinterval, makinga clarificationnecessary,is that
curial duty (cf. above, p. 29). capita in the law of 428 had been erroneously interpreted by some administrators
Constantines letterisparticularlyusefulforinterpretuigthepapyrusof3 59where (perhapsmunicipal) to refernot to the old sharesofassessmentbutratherto a distinct
'for the first time the unit oflanded taxation is no longer the concrete [aroura], assessment on heads, which was added on to the wgatw tenena.
the standard [Egyptian] unit of surface, but an abstract unit, kephale" (Deleage,
Capitation, p. 114). The papyrus in question (ibid. pp. 112-13) lists eight taxes with n.71 For example, Anastasius is saidto have abolished something ofthis sort (John
 134 CapU t AND COLONATE APPENDIX II: SUPPLEMENTARY NOTES 135

LydusDemagist.3.47); CJ11.48.23. pr. refersto thecapitahslllatio ofcohniadscripticii; n. ii By the sixthcentury, andperhapsalreadyin 396, lawsforbiddingsuperexactio
and one must account for the Middle Byzantine kapnikon. The western chevage is relate to taxes; in 325, however, the referenceis clearly to rents. May it be inferred
likely to have had a late Roman source (Marc Bloch's denial of this, Melangeshis- that taxes, which had once existed side by side with rents, had swallowed up and
toriques, I, 447 with n. 2, was hasty). What we do discern from the late fourth century superseded rents?
onwards is the divorce of this element from the main system of imperial assessment,
thatis,fromthesysteminwhicha caputhadbeena shareofassessmentequivalentto a n. 28 The workofNorbertBrockmeyer, ArbeitsorganisationundSkonomischesDenken
wgum. Hereafter, assessment of this sort was exclusively landed; the only link it
retainedwiththecaputofthepastconsistedin theGreektermziigokephale,for whose
inderGutswirtschaftdesr'omischenReiches(Bochum,1968),cametomyattentionwhen
meaning see below, ch. vi, nn. 12, 15. Capitatio humana proceeded rather from the
my bookwasinpress.Likemanyothers, Brockmeyer mistakenly regardedthebound
colonate as already existing in 332, but his research is principally concernedwith
figure, wholly unknown to us, at which the unadomed human head had been valued
in the formula census, and this, presumably, was much less than the yield value of a
earlierperiodsofRomanhistory, aboutwhichhearrivedatusefulandcomparatively
novel conclusions. The most interesting ofthese is that "the partial transition from
iugum. My guess is that the government generally abandoned this "head tax" to
landlords (including itself) as an item of rent.
an economic system resting on slavery to one based on leasing [to tenants], as it
emerges [from Cato, Varro,andColumella], hadits causenot in thedeficientpro-
CHAPTER V
ductivity ofslavelabour but in the changing economic thinking ofthelandowners
(p. 289),whopreferred to collect rentsasabsenteesratherthanto occupythemselves
n. 5 CJ 7. 32. 5 (ca. 291), redress against a tenant who has sold his landlord's land;
directlywithestatemanagement.Thisisa significantdeparturefromtheinterpreta-
4. IO. II (294)> a landlord is not responsible for his tenant's personal debts; CTh 11.7.2
tionsthathaveprevailedhithertoanddeservesto beextendedto theanalysisoflate
(319) shows a tenant paying taxes in behalfofhis landlord (above, ch. iv, n. 3); 4.22.1 imperialevidence.
(326), a tenant in a list ofpersons who might negligently guard one's property. Two
passages of the Digest have been thought to document the binding ofcoloni prior to n. 32 Oneshouldquestionthepersistentmodemconceptionthatthecolonatebrought
thefourthcentury (cf. Clausing,Colonate,p.279n. 5):theonediscussedabove,ch.rv, abouta rapprochementbetweenlandlordsandcultivators.Forexample.MarcBloch
n. 4 , and Dig. 30.1. 112. pr. (Marcian): "Si quis inquilinos sine praediis quibus in Camb. Econ. Hist., I2,257, " ... the binding ofthecolonus reinforced most eiFectively
adhaerent legaverit, inutile est legatum: sed an aestimatio debea.tur, ex voluntate thedependence ofsmalllandholders ontheirlord"; it wouldtendto transform those
defuncti statuendum [est] ... " One may indeed draw "(house)-tenants who adhere to affected "into a disciplined group, exceedingly stable, in which the employer s
thefields fromthispassage(asoriginallydoneby Cujasin 1577:Clausing,Colonate, authority over men who could only get a living on his premises would be greatly
p. 31), but upon reflection such an interpretation seems absurd. The intent of the increased. " One might infer this from the confinement of coloni, but that fact was
passage is to prevent the bequest ofrental property apart from the soil going with it, offset by the continued remoteness oflandlords, whose absenteeism showedno per-
includingthat on which it stood, while still respecting the wishes of the deceased ceptible decrease. We find Symmachus saying {Epist. 9. 6), "actores absentium, quibus
by permitting compensationfor the value ofthe object bequeathed.Boundcolonior res longinqua committitur, tamquam soluti legibus vivunt, " as well as calling on
mquihnicannotbeuncoveredin thesecondcentury onthebasisofevidenceasflimsy public authorities for assistance (5. 87;7. 126;9. 40and 140) orcomplaining about thenr
as this. (7. 66, "ut fieri per dominomm absentiam solet"; 9. 10) For dramatic evidence of
I leaveout ofaccountthelawsregardingimperialcoloni,on thegroundsthat, in absenteeism in sixth-century Amida, see Joshua the Stylite Chronicle 83-4, ed. W.
legislatingforthetenantsofitssoil,thegovernmentwasinthepositionofa sovereign, Wright(Cambridge,1882),p. 66;andItaly,ProcopiusWars7.6.5-7, I3-I.22.20-21.
nota landlord.It isnoteworthyin thisrespectthattheearliestappearanceoftheterm The notorious phenomenon ofpatrocinium, which is often taken to document the
iuscolonatusin theCodes,CTh12. 1.33 (342),relatesnotto a disabilitybutto anabuse mediatizationofpeasants,is betterenvisagedasreplacingonetype oflandlordby a
bywhichmenofsubstanceavoidednominationto thecuriabyrentingsomeimperial new type, without substantial change in the condition of the culrivators, except,
landandthusqualifyingformunicipalimmunity, i.e.,CJ11.68. 1 (3255). Forcontrast probably, a decreasein their payments. (For a newandhighly refreshing assessment
seethesenseofiuscolonatusin CTh5.6. 3 (409).It isnotsurprising,therefore,thatthe ofpatrocinium, seeP.R.L. Brown, "The RiseandFunction ofthe Holy Man m Late
emperors should have legislated on the duty of their coloni to engage in cultivation: Antiquity, " Journal of Roman Studies, 61 [1971], 85-7. ) Given the dominant fact of
CJ 11.68.2 (319), h.t. 3 (3648), 7. 38. 1 (367$); on the marriage of their wlonae: CTh absenteeownership,I suggestthatthelawsonthecolonatehadpreciselytheopposite
4. 12. 3 (32os), CJ 10. 32. 2p (365), 6.4.2 (367). 11.68. 4 (367); onjurisdiction over them: effect to that envisaged by Bloch. Their result, if not their object, was to substitute
CJ 3.26.7 (349), h. t. 8 (358), CTh 11. 7. 11 (365), 10.4. 3 (373); on their dues and taxes: public authority for that ofprivate law in the owner-tenant relationship. Ideally, the
CJ 11.63. 1 (319), CTh ii. 17. 1, io. i. ii = 12. 6. 14 (367), 13. 1. 8 (370), h.t. io (374), lawswould havesafeguarded landlord incomes while leaving landlords freeto pursue
CJ 11.68.5 (386$); andon their freedomfrom eviction: CJ 11.63.3 (383). their urban careers - but this at the expense of any sort of community of interest
 136 CapU t AND COLONATB
APPENDIX II: SUPPLEMENTARY NOTES 137
with their tenants, whom public law alone was expected to keep at their tasks. The baking bread, limebuming, road and bridge repairs, and other tasks ofthis sort. It is
subject requires closer investigation. worth stressing, however, that the earliest text I have found which uses this adjective
reads as follows: Dig. 50. 1. 17. 7 (Papinian), "Exigendi tributi munus inter sordida
n. 50 Jones, LRE, p. 1043, concluded "that the population dwindled because when munera non habetur et ideo decurionibus quoque mandatur. " The task mentioned
theyhadpaidtheirrentandtaxesandotherexactions,thepeasantryhadnotenough here, though unattractive, ishardly comparable to those listed in 382 (cf. the parallel
to rear sufficient children to counter-balance the very high death-rate. " Jones had law of 362, cited on p. 38), and Papinian's intention appears to be to distinguish
already stated this thesis in Ancient Economic History (Inaug. Lect., London, 1948), between occupations suitable and unsuitable to the honour of decurions. Since a
pp. 17-18, and maintained it in N. Cantor, Perspectives on the European Past: Con- common denominator between Papinian (ca. 200 A. D. ) and the late fourth-century
versations with Historians (N. Y., 1971), pp. 123-4; he evidently regarded it asvery im- laws is very difficult to discern, we are in a poor position to trace the historical
portant. If, however, theonsetoftheboundcolonatehasbeencorrectly interpreted development ofthis expression. The significant step, to my mind, is documented by
here,theideathatpeasantspaidbothrentandtaxesafterthe3705orsoseemsexcluded. CTh 11. 16. 5 (343) and consists in the identification of sordida munera with munera
It isdifficult to see,inanycase,howrents (over andabove taxes) could havesubsisted extraordinaria, an equivalence further illustrated by the title headingof CTh 11. 16,
ina situationwherelandforcultivationwasapparentlyoverabundant (below,n.69). 'De extraordinariis sive sordidis muneribus"; cf. CTh 6.35.4 (321), as a possibly
Thehypothesis thattheeconomicdeclineoftheEmpirestemmed fromdiminishing preliminarystep in thisdirection. Jones,LRE,pp. 535-6, assertedin commentary to
pressure upon producers seems preferable to one attributing decline to their having the laws of 382 and 390 that they distinguished extraordinaria, which everyone had to
beenoverexploited. perform, from sordida, which a specified list ofdignitaries were exempted from; but
this interpretation presupposes that the legislator kept the categories distinct rather
n. 54 Thefirstform ofcommutation isattestedfrom soearlya date (Cerati, p. 246, than, as I believe, using them as synonyms. In any case, Valentinian m (446) had in
andabove ch. m, n. 2*) thatit shouldberegardedasa.constant accompaniment of mind something like the list of 382 when he complained that the invention of the
levies in kind; cf. theinteresting observations ofW. Liebeschutz, "Money Economy phrase munera sordidahad resulted in denying the state everything it needed to defend
and Taxation in Kind in Syria in the Fourth Century, " Rhemisches Museum, 104 the safety andhonour ofthe Empire (NValent 10. 1.3); the adjective hadbeen a spring-
(1961), 242-57, esp. 254-5. The more important phenomenon oftranslatio to gold board to massive evasion.
payments isdocumented bya changeintheterminology ofthelaws,inwhichannona
beginsby beingthe genericterm for all taxation (e. g., CJ 11. 55. 1 andthe Brigetio n. 29 la Epist. 47, he cites the total as more than 62, 000wga ; in Epist. 42, he men-
Table)andthenturnsbythe3 8osandthereafterintoonlya partofthetaxlevy (e. g., tions 10,000 tamiaca (= res privata) and 50, 000 eleutherica (= free, see below) wga,
CT\ ". 1-19, 11.2.4, II.152 - forming a single law). Cerati attributed the change of which 15, 000 used to be chrysotelon, gold paying, to the collectors of the comes,
chieflyto theexigencies ofgovernment; themilitary requirements ofthestateafter presumably the comes sacrarum largitwnum (ed. Azema, as above, n. 10, pp. 123-5,
Adrianople called for gold rather than quartermaster stores. Another and perhaps iio-ii). He also describes a transaction that took place in 435-6, by which the
complementary approach to anexplanation might contrast thevariable taxationthat comitiad exchanged 2, 500 iuga oftheir lot with the Pretorian Prefect, on the grounds
waspossible with levies in kind (i.e., basedonproduction) to thefixedtaxation im- that these iuga were barren. Jones, LRE, p. 816, discussed this transaction under the
plicit in a.tax in gold (i.e., basedon eammgs). Ifconversion to gold seta limit on heading of agri deserti, but this overlooks the fact that an exchange is taking place:
tax increases, it would no doubthavebeenpopular with thetaxpayers. the wgaare barren only because they are no longer able to bear a tax in gold; they
can still be solvent, however, vis-a-vis the Prefect, who exchanges them for wg&from
n. 56 The sameinterpretation should begivento CTh 13.4.4 (374), "Picturaepro- his lot that arecapableofpayinggold but maynot formerly havedone so. Whatis
fessores, si modo ingenui sunt, placuit neque sui capitis censione neque uxomm aut described, in other words, is simply a translatio praestationis.
etiam liberorum nomine tributis esse munificos et ne servos quidem barbaros in I. Hahn, "Theodoretus Cyrus und die friihbyzantinische Besteuerung, " Acta
censuali adscribtioneprofiteri"; here, the distinct reference to tribute, makes it even antiqua, 10 (1962), 123-30, was right to reject the interpretation of eleiithericon as
clearer that censio or censualis adscribtio was a matter of assessment. The Palestinian "immune"; cf. Scheltema, An denWwzdn, p. 143. He got into difEculties, however,
clergy and African painters would, unless exempted, have been entered on local by identifying the wgapertaining to the resprivata with those paying in gold, and by
assessment lists, thus incurring anobligation to taxesandmunera ofsomekind; but the failing to note that gold payments to the largitiones for leased imperial land are com-
lawsdonotallowusto specifytheobligationsfrom whichtheywereexempted. monly attested in the laws.
CHAPTER VI
APPENDIX I

n. 3 Regarding munera sordida, the most comprehensive list appears in CTh 11. 16. 15, a. 2 Deleagereasonedas follows: a place called Iug{era)qwnquagmtais rated at four
18 (382, 390), western and eastern editions ofthe same law. The munera include m{iUenae};hence,a milknaconsistedoftwelve anda halfi'Mgen.The flawhere, quite
 138 Capl it AND COLONATB

apartfromtheresolution ofm,istoassumethata placenamedafteritssizeoncelong

agoretainedprecisely thatsizeinthefuture. Seeck, "Schatzungsordnung, " pp. 303-4,
wasalsocertainoftheresolution millena. Sinceplacesprefixed K, forcasa,wererated,
he inferred that the millena assessed persons along with land, on the grounds that
casaewerehutsandthatinhabitantsweretheonly thingabouthutsthatwasworth
assessing.This too is a somewhatreckless interpretation ofplacenames. Schwahn, Postscript
Tributum, " col. 71, listed the inscription as the earliest evidence for the millena.
J. Susta, "Zur Geschichte andKritik derUrbarialaufzeichungen, " Sitzungsberichte der
kais. Akad.derWiss. {zuWien),philol. -hist. CL, 138(1897), no. 8,p. 6 n. 4, took the
inscription out ofconsideration. Segre, "Studies, " p. 112n. 57, rejected the reading
millena because none ofthe entries include fractions (an excellent reason); he did not
explain,however,whyhealsorejectedthealternativeofmodn.

A footnote to chapteriv containsthestatement that "little would be lost if

new evidence proved the existence ofthe assessment wgum earlier in the third
century than Diocletian's reign. "1 Other than associating the iugum with
Diocletian and entertaining the possibility ofan earlier origin, the preceding
chapters did not seekto establishwherethe iugummight havecome from;
there seemed to be no evidence on which such a discussion could be firmly
based. Since then, I have found sources that suggest the probable origins of
the term and offer a more concrete pattern for the assessmentschemethan I
was able to reconstruct from legal evidence alone.
All commentators, myselfincluded, have believed that the term iugum
asusedin lateRoman taxation wasrelated,more or lessabstractly, to a "yoke
ofoxen,theimplicationbeingthattheiugumhadto dowithagriculture,the
obvious basis ofeconomic life. 2 A very different derivation for iugum is pro-
vided by the Tactica of Aelian and Arrian, Greek military authors who
worked for the emperors Trajan and Hadrian in the early second century.
In their writings, zugon means a "rank of soldiers, especially the foremost
rank.3

i Ch. w, n. 19.
2 For a very forthright statement, see Lot, ImpStfon der, p. 2 n. 8, who quoted the equally
forthright Godefroy: "e^ojugaputem dictaterraemodum cui colendo per annumjugo
boum opus est. " After a period during which scholars hesitated, the yoke was brought
back to the fore by Zachariae von Lingenthal: Deleage, Capitation, p. 9.
3 Liddell and Scott, Creek-English Lexicon, ninth cd. (Oxford, 1940), s.v. zugon vm,
rank or line of soldiers, opp. file (stoichos). " The usage dates from as early asThucydides,
along with many other meanings, including "yoke" for a carriage or for oxen; the
underlyingideaappearsto bethat ofa pair.The specificsenseof"frontrank"is notedin
Aelian Tactics 7. 1 and Arrian Tactica 8. 1. That they - the contemporaries ofHadrian -
should embody this distinction between the front rank and those behind is highly signifi-
cant to taxation, as will be seen; it may also be important to studies of the late Roman
 140 CapU t AND COLONATB POSTSCRIPT 141

To interpret this clue, one must understand the sense of "rank" and particular square illustrated above is formed when eight times eight men are
'file" in the literature on tactics. The lowest subdivision ofan army consists lined up [stoichein') and their ranks arejoined {zugdn\
of the file, which is composed of a variable but even number of men. Such is the terminology offered by the tacticians for the rank and
Arrian favoured a file (usually called lochos) of eight men. 4 In a simplified file" and for "liningup" soldiersand "joining"them. This wasthe pattern
diagram, the file would appear in the form ofa leader with seven men behind thatthe government applied to taxassessment atthe close ofthe third century.
him. Such files are the building blocks oflarger tactical units. For example, The tax reform known from the edict of 297 was carried out by a regime
one might form a tactical square ofeight files, totalling sixty-four men, asin apt ly described by Peter Brown as "the highly effective and unashamedly
the following diagram (the symbols T and o are those used in the diagrams military junta that we dignify by the grandiose title of'The Diocletianic
that accompany the Aelianic corpus):6 Tetrarchy.' "7 That the Tetrarchy should have been inspired by military
models in its regulations of civilian life has often been intimated; a graphic
TTTTTTTT illustration is provided by the house on the plan of a Roman fort that Dio-
00000000 cletian built for his retirement. 8 It is hardly surprising, therefore, that two
00000000 semi-official writers on tactics were the source ofinspiration for the assess-
00000000 ment scheme instituted by the Tetrarchy. What remains to be seen is how
00000000 the tacticians' model was translated into the civilian sphere to produce the
00000000 Latin terminology o£iugatio and capitatio.
00000000 The application ofthe model to the civilian population presupposed
00000000 that,just as the army had its front-rank men (principales) and its ordinary
soldiers, so could civilians be divided into leaders and followers. In fact, this
The square therefore consists of eight "ranks, " zugoi, and eight "files," bisection ofthe population had been long prepared by Romanjurisprudence,
lochoi. notablyin criminallaw, wheredifferentpenaltieswerelaid down for hones-
To these nouns there correspond two verbs that, in conjunction, pro- tiores and for humiliores. The bisecdon ofthe population into "rank and file'
duce the square. The first one, stoichein, means to "be in line, " the line rele- would continue to be reflected in the many laws ofthe fourth century and
vant in this context being a stoichos, "row in an ascending series, " that is, a later that distinguish potentiores from tenuiores, mediocres, or infimi.9
military "file {lochos to Aelian and Arrian). The second verb, zugein, is The next step in a regulation about taxassessment [census] wasto draw
cognate with English "join, " usedherein the senseof"joining ranks. "6The up a set ofequivalences,a formula specifyingtheminimum property rating
ofa civilian front-ranker; this schedule ofassets - a "joining ofthe various
army. OntheGreektacticians,seeAlphonseDainandJ.A. deFoucault,"Lesstrategistes elements ofwealth - defined the fiscal iugum, as we have seen it in the Syro-
byzantins," Travciux etmemoires, 2 (1967), 315-92.
4 In a lochos oflight-armed troops: Arrian Tact. 4. 3, following Asclepiodotus Tactica 2.7
(first century B. C. ). Aelian and Arrian double the number for heavy-armed men: Liddell grateful to Dr. Michael Jeffreys, Visiting Fellow, Dumbarton Oaks, for his help with
and Scott, Lexicon, s.v. "lochos." these terms.
5 On the diagrams, which apparently stem from the Tactics of Asclepiodotus, see 7 P.R.L. Brown, "Approaches to the Religious Crisis of the Third Century A.D.,"
A.Dain,Histoiredutexted'ElienIeTacticien(Paris,1946),pp.65-8,whospecified(p. 66 EnglishHistoricalReview, 83 (1968), 555.
n. 3) that the symbols, in lower-case letters, are conventional signs, not initials. It is 8 Ernest Hebrard and Jacques Zeiller, Spalato: Le palais lie Diocletien (Paris, 1912),
curious to observe that, in the Rangordnung oftheearlyByzantine Strategikon of"Maurice," pp. 26, 53, and 20 with n. 2: a castelhim in our eyes, but called villa by historians of the
the lowest elements are called tetrarcha and oragos (1. 3); the same passage also identifies secondhalfofthe fourth century. In his pioneeringstudyofSpalato,RobertAdambased
thislowest elementwitbphylakes,aboutwhichseeGoffart, "From Rom. Taxat.to Med. his interpretation of the rums on the plan of villae supplied by Pliny the Younger and
Seign., p. 381 with nn. 137 and 157. Vitruvius: Ruins ofthe Palaceof the Emperor Diocktian (London, 1764), pp. 7-17.
6 Liddell and Scott, Lexicon, ss.vv. "stoicheo, " "stoichos, " and "zugeo. " Pollux the 9 Peter Garnsey, Social Status and Legal Privilege in the Roman Empire (Oxford, 1970),
Grammarian(secondcenturyA.D.) applieszugonandstoichostothechorusina play.I am pp. 231-3 and passim.
 142 CapU t AND COLONATB POSTSCBIPT 143

Roman Lawbook. 10 Any man whose declaration of property (professio), fessiones ofwga ("rank") to be distinguished from the capitationes of\uga
when sifted through t\iefor mula produced
census, one iugum or more was by ("file"),asina lawof340;"andonemightspeakoftheiugatiooftherichas
the same token a citizen ofthe fi-ont rank. By a very simple transference of distinctfi-omthecapitatiothatgroupedpoorermeninto "files"oftaxpayers.
ideas, one may seehow a declaration meeting this minimum could be called An earlyreaderofAmmianuswouldhaveunderstoodthatthewayin which
a caput,thatisto say,a "heading"or, more specifically, a "fi-ont-rankhead." the emperorJulianhadloweredtaxesin Gaulhadbeento renewthe census
Citizens having assets below the minimum were more difficult to andto increase the number ofcapita, "shares ofassessment, to such an extent
assess; they hadto begrouped into "files, " clubs oftaxpayers - what had been that each capitulum, "file" or "club" of taxpayers, had to pay only seven
named a "symmory" in ancient Greece and was called capitulum in Latin." solidiper year insteadofthe twenty-fivesolidithat were exacted before. As
In fact, capitulum is precisely the word used by Ammianus Marcellinus in a was to be expected, the tendency ofthe tax rolls, in theyearsafter anassess-
passage regarding taxation that has previously resisted interpretation. 12The ment had taken place, was to fall away from the fragile "files and towards
process bywhich these citizens were assessed, or formed into files, wascalled the solid "front ranks";16as the tax baseshrank, the amount demandedfrom
capitatio; again, the official schedule ofassets, orformula census - one schedule theremainingcapitulatendedto rise.Andjustasa goodgeneralfilledoutthe
for realproperty, onefor "animate" assets- wasbrought into play, andwith depleted numbers ofhis units by recruiting additionalmanpower, so did a
itshelp, onefamily ormore oftenuiores wasconverted intoa single"file" or good civilian administrator set in motion the assessorsto fill out the files
caput of tax assessment, subject to the same contributions to taxation as a oftaxpayers, in order to relieve the overburdened front ranks. 17
'front rank" (fugum) ofproperty. 13 So much for a briefexposition ofhow the language and methods of
Oncethesecensus operations hadtaken place, the total assessment of a the military tacticians provided a model for Diocletian s reform oftaxation.
city or other district could be expressed in multiples ofhomogeneous units,
called capita; a less precise though still possible collective term for these 15 Cf. above, ch. m, n. 36.
16 The reference to the years after an initial assessment allows me to comment somewhat
multiples wasiuga,whilethemostaccurateofallwastheGreekcompound more firmly than before on the troubled question whether assessments were periodically
zugokephalai, whichhadimbedded in it theideathattheassessment ofiuga. renewed(above, pp. 108-9).Thetraditionalmodern view appearsto be that, although
ofindividual declarants wasaddedto theassessment bycapitaformed ofclubs evidence is scanty, renewals did take place at definite intervals: Stein, Bas-Empire, I,
74-5; Svoronos, "Recherches sur Ie cadastre byzantin, " pp. 63-7. An interesting alter-
oflessercontributors to taxation. 14Theassessment schemeallowed thepro- native is suggested by M. W. E. Glautier, A Study in the Development of Accounting in
Roman Times, " Revue internationale des droits deI Antiquite, yd ser., 19 (1972), 314, where
io Asabove, ch.m. Inlight ofthisnewinformation, it seemsprobable thattheequiva- the "passive" accounting of Antiquity, which merely offered a check upon ongoing
fenceslaiddownby the Lawbookportray the iugumasit waswhentheLawbookwas action (usus, consuetudo), is contrasted to modern "dynamic" accounting, which keeps
compiled (ca. 500), rather than under Diocletian, regardless of the implication to the continual track ofwhat is going on (i. e., gain or loss) and is itself thereby an instrument
contrary;theiugumdefinedbys.formulacensusofca.300islikelytohavebeenmuchlarger. ofpublic administrationor private business. In the perspective of passive accounting,
Inkeepingwiththemilitary originsofthescheme, I would guessthattheoriginaliugum it would seem to follow that the record of a formal census served the same purpose as a
approximated theNorthAfricancentmia (againa term with a military resonance), aswell boundarystoneinthefieldsor a painstakingcenturiation:it servedasa baseline- notthat
astheverylargeGallicancaputmentioned inSidonius'spoem of459(above, ch.iv, n.26*, boundaries or a census were immutable, but in order that there might be a check on the
revising what is saidthere). legality ofthe mutations that subsequentlytook place.This is why one might conceive
ll For symmories, " see, for example, Andreades, Greek Public Finance, pp. 324-5. For of carving a census on stone (see above, p. 121), as well as why only one P. Ital. 3, one
capitulum TLLm, coll. 350-52,esp.parag.2; thelateRomantaxusageseemsto derive Polyptych ofSt. GermaindesPres, oneDomesdayBook, andsoforth, are in existence.
from theideaofthe"capital"asthetopofa column (cf.themilitary filewithitscapuf), Renewals of assessment, therefore, were the exception, not the rule, much less a periodic
used in Vitruvius and Pliny the Elder. The normal context for capitulum in CTh is process; ifthe ideallegal order endured, no renewalwouldever be needed.
recruitment; see above, p. 25 {capitularius). 17 Asabove,n. 12.Thebestclueto aninterpretationofthepassage,otherthanthedefini-
12 Rer. gest. 16. 5. 14.The interpretationis in thenextparagraphwithn. 17. tion ofcapitulum, is offeredby Amm. 19. 11.3, explainingthe splendidwork ofAnatolius
13 ThisismuchthesameprocessasI morehesitantlyinferredfromtheAutunPanegyric as Praetorian Prefect oflllyricum: he relieved titubantes by revising the assessment rolls
(ch. re). Capitatiointhissensesolvestheproblemencounteredabove,ch. m, n. 15 and {censiialisprofessio). In Amm. 17.3.2, 4, capitatio means ordinary taxation. Ammianus
ch. iv, n. 11, regarding Dig. 50. 4. 18. 8 and 29. believed that Julian's enthusiasm for "filling ranks" sometimes had undesirable effects,
14 Cf. above, p. 96. notably in the case of curiaks: 22. 9. 8, iz; 25. 4. 21.
 144 CapUt AND COLONATB POSTSCRIPT 145

The scheme ofassessment based on "rank and file" closely agrees with the are the aspect ofinstitutions that get transmitted by being adaptedto new
laws that have been analysed in chapters m and iv. The contention that early conditions. Naturally, the implications ofthis observation - which is hardly
fourth-century assessment was based on persons and not on land benefits profound - will haveto bedeveloped elsewhere andat greater length. But one
from additional support. We can now see that the very notion ofthe wgum item ofevidence may be cited to illustrate the point. We have seen that the
wasjust as personal as that ofthe caput, and that the process called capitatio lochos, file, ofAelianandArrianwassubjectto being "linedup, stoichein;
consisted of forming "files" rather than of counting heads. The Greek in a sixth-century Byzantine tactical treatise, we find that lochos is replaced
tacricians had come to the assistance ofthe Roman surveyor, superseding the by an older term for "file, Xenophon's word stichos, a. noun related to
laborious delimitation of land by the expeditions brigading of taxpayers in the verb; and stichos occurs in an anonymous tenth-century tract on tax
proportion with their declared assets. 18 By this translation ofmilitary con- assessment,aswellasinaneleventh-centurytaxrecord,asthebasic"heading"
cepts to the municipal sphere, the emperors succeeded in sustaining and for purposes oftaxation- in short, the early fourth-centurycaputin Middle
perhapseven strengtheningthe traditionalideathat taxationwasthe active Byzantine dress.21 Diocletian had built a solid edifice, one whose solidity
functioofmenratherthana passivepraestatiofor property. FrancescoGrelle's resided not in its immediate fabric but in the techniques by which it wasbuilt.
conclusion on this subject, which he applied to the second and third centuries Thisbriefpostscriptshouldclosewitha comment on the socialimpli-
on the basis ofthe Digest and which I extended into the fourth century on cations ofDiocletians taxreform. Assessment by"rank andfile"wasacreative
the basis ofthe Codes, has now acquired the support ofa new category of novelty, well designed to draw together the constituent parts ofthe body
evidence. 19 politicandtoremedytheirsocialdivisions;it wasnotjusta wayto copewith
Another point that the military authors confirm is that iugatio- social abuses. 22 The present book has often drawn attention to the fact that
capitatiowasa systemofassessmentandnota tax.A studyoftaxation,unless Classical Antiquity traditionally demanded that the rich should contribute
it is concerned only with short-term developments, should addressitselfto more to the community than the poor, and it hasstressed that this tradition
processes rather than to the fixed forms that these processes acquire from was prolonged into the later Roman Empire. The object ofstatecraftwas to
time: taxes come and go, as occasion dictates; what endures are systems of restrict individualism to the extent necessary to attain social cohesion, and
assessment.Ontheotherhand,processesarealsosubjectto changeor at least themilitarymenwhoreformedtheEmpirewerewellawarethatthestrength
to reform; it is clear that assessment by "rank and file" was a novelty of ofa state was founded on the concord ofleaders with followers. Just asin the
far-reaching importance. Its endurance cannot be measured by the specific 'square" ofan army the file andrank were interdependent, with the front
terminology ofiugatio-capitatio, which was in full use for barely a century. rank enduring greater risk than the rear ones, so in the civic sphere were
The enduring aspect ofthe reform was not this specific constellation but the
process itself: the minimum standard for principales and the "clubbing" of r34-5 (no- 48). Byzantium: the legislation on military lands, presented by Lemerle,
tenuioresintogroupscapableofattainingthisminimum. Systemsofthiskind, Esquisse, in Revuehistorique,220 (1958),43-54;thepassageofTheophanesdiscussedby
for a variety ofpurposes, arevisiblenot only in the fourth century but also Lemerie in ibid., 219 (1958), 72 (I interpret it somewhat differently) implies that the
emperor Nicephorus (802-811) was forcing the "rank" to reconstitute the "files, " just
in the empire of Charlemagne, in tenth-century Byzantium, and perhaps asin thefourth century largerlandownershadbeensubjectto collatio tironum. Asregards
even in the Angevin Empire ofHenry n. 20This was no accident: processes Angevin England, I am tentative because the necessary books are momentarily un-
available to me; I have in mind the Assize of Arms, but this may be an exaggeration.
2i Anon. De re strategica 15. 9-13, in H. Kochly and W. Rustow, Grieschiche Kriegs-
l8 Themeeting oftactician andsurveyor isnotjust a metaphor. Seewhat issaidabout the schriftsteller, r. 2, Abt. 2 (Leipzig, 1855), 94-5; cf. Liddell and Scott, Lexicon, s.v. "stichos,"
centmia above, n. 10. The evidence from Syria about the delimitation of city territories citing Xenophon Resp. Lacadaem. 11. 5. 8 andDe equit. magist. 3.9. Interestingly enough,
under Diocletian (above, ch. rv, n. 9*) suggests that the land surveyor was relegated to the Anonymous anticipates Godefroy (above, n. 2) by underlining the derivation of
dealing with larger territorial boundaries while the tactidan's method was applied to zugonfrom a yokeofoxen (15.9): hereiswherethecrucialstepfrom thetaxwgumto the
government surveys within tax districts. seigneurial terra uniusfamiliae ("hide" or mansus) is first spelled out - not in a western but
19 For Grelle, see above, ch. m, n. 21. in a Byzantine source (cf. Goffart, "From Rom. Taxat. to Med. Seign., " part i).
20 Fourth century: e. g., CTh 7. 13. 7 (375). Charlemagne: Man. Germ. hist., Capitularia i, 22 Above, ch. m, n. 10* and ch. v, n. 59.
 146 CapU t AND COI.ONATB

potentiores and tenuiores obligated to one another; the cure for social ills, as
envisionedby thelateimperialgenerals,wasto reform the "square," whose
cohesion was as much the strength of a political community as it was the
mainstay of a disciplined army. The city of Rome had become great by
translating a body politic into military force. It is little wonder, therefore,
Indexes
that the leaders ofthe mature RomanEmpirederiveda model for imperial
renewalfrom the military roots ofits pastgreatness.

Dumbarton Oaks, Washington, D. C.

February 1974

I INDEX OF SOURCES
The sources are indexed by chapter (in Roman numerals) and footnote number under three
headings:LawCodes; Other Legal;Literary. Notes markedwith an asterisk areto be found in
Appendix u (beginning on page 126). The location of texts marked with a dagger is given on
p. 151.

LAW CODES io. 32. io/imi8 11. 63. 1/TU5*

io. 32. i3/nn22 ll. 63.3/vnn5*, 60
CODEX JUSTINIANUS I0. 32. 29/vn5* ll. 68. l/im24; vn5*
l. 3. l6/vn49 10.42/imS ll. 68.2-4/vn5*
1.55.4/vnil 10.43 .i/mn4 11.68. 5/vnn.s *, 10
3.26. 7-8/vn5* io. 55.2-3/nni4* I2.52. I/FVU56
4.9-I/IIIU9 II.38.i/imi4
4. lo. ll/vnn5*, 33 11.48. 6/vnni4, 34, 5l> 55
4.46-T/mn9 li. 48. 7/vnn35-6, 39-40, 51, CODEX THEODOSIANUS
4.46.2/11104 60; vm39 l. 28.2/mn2X
4.49-9/lTni6> 21, 36 11. 48. 8/vnn33, 35. 42 3. i.2/ivn6i; viryS
4.65. 5/VU34 ll. 4S. l2/vn48 4. l2. 3/vnn5*, 14
4. 65. I7/VU33 ll. 48. l3/vnn40, 48 4.22. i/vnn5*, l6
6. I.4/VU48 II.48. 2I/VU24 5.2. I/IIU22
6. I.7/VU48 ll. 48. 23/ivn7l*; vnil 5. 6. 3/vnn5*, 40
6.4.2/VU5* ii.^o.T/vnii 5. il. 8/mn7*
6. 3I. I/IVD4* ii. 50. i/vnn6, 60 5. l2. 3/vn69
7. 30. I/VQ34 11. 50.2/vnnii, 18, 24-5, 49 5. i3. i/inni3
7. 32. 5/vnn5*, 34 ll. 5l. l/vnn56-7 5. i3.4/lT2l; mi4l
7. 38. l/vri5* ll. 52. l/nmai5*, 22; vnniS, 5. l6. 34-5/vn69
7. 39.a/vn34 24, 44, 47> 62; vmi6 5. l7-i/m"2o; ivnyi; vnni2,
7. 62.7/imnS, 18 II. 53. l/vnn23, 35, 43. 51 39, 5l, 58, 60; VIU39
7. 64. 3/vm25 11.55. 1/111125;Ilinzo; ivnnfi, 5. I7.2/VD47
7.64.9/IIUI4* 17, 20, 22, 64; vn54* 5. I7.3/VU42
9.2I/VIU25 II.59-I/IU28; im23; VD4 5. I8. I.2/VD46
lo. io. i/nnza ll. 59. ll/vnn4, 6g 5. I9. I/VU34
io. l6. 3/mn4* ii.62.2/mn33 6. 3. 3/VIU40
io.27.a/vmi2 ll. 62. 5/vnn67, 69 6. I3. I/IVU48
 148 INDEX OF SOURCES INDEX OF SOURCES 149

6.2I.2/IVQ48 ll. 7. 2/ivn3; vn5* I2. I.33/VU5* 33.2.28/ID27; vni6 NMAJ EDICT OF 297
6. 22. 1/11022; IIID25 ii.7.4/vn69 l2. l. 36/imu6 33. 2. 32. 9/m27 7. 1.i6/Append. mi see P. Cair. Isidor. I
6.22. 2/II172 ll. 7. ll/am34; vnj* l2. l. l55/vn69 43. 32.l/ivn4* 7. l. l7/vin46
6. 35-3/im"l5, 22; ivnn3i, 58 ll. 7. l2/m50; vm40 l2. l. l6o/vn69 43. 33. I/VU33 EDICT. THEODOR.
6. 35. 4/VB13* II.9/IIIU9 l2. l. i6i/vn69 48. 19. 10.pr. /vn20 NTHEOD
I42/VU40
6. 35. lo/mi5I il.9.2/ivn6 l2. l. l8l/vn69 49. l4-2l/nm9 7. 4. 2. 5/TO49 I49/VII146
7. l. 3/ivnn3l, 58 ll. lo. l-2/ivn2l l2. 4. l/mni5*; rvn67* 49-l4-28/mn9 22.2. i2/mni5*; ivn67
7.4. I/IIDI4 il. i2. i/ma34; ivnn30, 64 l2. 6. l4/vn5* 49.46. 9/vn33 PRAGM. DB IURE PISCI
7.4-32/ivn72; Vini5 ii. l2.2/mn2l; ivnn6, 41 I3. I. 8/VU5* 49. 47-J"'-/I""9 NVALENT
5/mn9
7. 5-2/IIII12I li. l3. 3/vm37 13. 1. 10/vnj* 50. l. l7. 7/vm3 5. 1. 4/mnii*; Append. mi
7. 6. 3/mnil; ivnny, 22*; lI. I5.2/Iin28; VU54* l3. 3. l/im5 50. l. 30/vni4 lo/mni5*; vmn3*, 33
FRAGM. VATIC.
VU56 il. l6.T/vm3* I3. 3. 2/III14 50.2. 1/imiS l3/ivn63; vm45
7. l3. 6/ivn52 ll. l6. 3/lin28; vnp l3. 4-4/T"5 ; vn56* 50. 4. 1.li/imio 24/VIU46 34/IVU3
7. X3. 7/mnn2I> 36 . IVD53 ll. i6.4/lin28; ivn26*; vn9; l3. 5. 5/im8* 50.4. 3-I/IVQ46 31. 1. 1, 6/vn46 35/ivn6l; vnn38, 60, 67;
7.2o.2/uni3 vm39
vm37 l3. 5. l4/vinn6, 34 50. 4. 3. 15/1109 35. l. l8/vn40
7.20. 3/imiz ll. l6. 5/vm3* i3. io. i/im28; ivni5 50. 4. 4. 2/VID3
7. 20.4/uni2; ivnn23, 42-4, ll. l6. 6/mn6; Iimi2; ivny; l3. lo. 2/mnni5*, 20; rvni6 50. 4. 18. 8/nmi5; ivnil LEG. VISIG. GAIUS

50.4-i8.26/m28; im23 12. 1.2/VniI Inst.

50 vmi5 l3. lo. 3/vnn29, 36, 39
7. 20. 8/IIUI2 ll. l6. 7/ivn38* l3. lo.4/lim28 50.4-l8.29/im8; nmi5 4. I53/VUI6
7. 22. 2/II1122 ll. lfi. S/rvryS; vm37 l3. lo. 5/vn67 <, o. 'j. l. pr. /imzS OTHER LEGAL
8. l. 3/mn2o ll. l6. i5/vm3* i3. io. 6/mna9 50. 5. I.2/IIU25; IVU2I HERO ALEX.

8. l. l2/vm40 ii. i6. i8/vm3 l3. lo. 7/vn67 50. 5. 3/nniS BRIGETIO TABLE Geomtr. l-ssid
8. ll. i/mi28; mnni3, 34; ll. l7. i/vn5* l3. ll. l/ivn63 50. 5.6/nn8 ivnn6, 40, 42, 44; vnn44, 54*
vn9 ll. 2o.6/mnni7, 22; ivnn55, l3. ll. 2/ivnn47, 56; vn63 50. 5. 9/imiS INSCR. Ami. DB SALTIB.
9. 3. 5/v"35 57; vn54; vmn9, 15 l3. li. l3/vn6g 50.5. lo.pr./im8 CASSIOD. vm4
9.42.7/IVU4* 11.22. l/m50; vn27 l5. l. 34Av"72; vmi5 50. 5. ll/im8 Variae
io. l. ll/vn5* II.22.2/m50 l6. 2. l/imi 50. 5. 15/liniS 2. 26/VB146 XNSCR. ASIAN. DE CENSU

I0.4. 3/vn5* II.22. 3/IU50; vm40 l6. 2. 2/imni, 6, 9 50. 15. 3/ivn47 2. 30/VIU46 IVUD5, 31, 59, 66; vni4;
lo. 8. 3/im22; vn3 li.22.4-5/ivn40 l6. 2. 3/im8* 50. i5.4/BnI4 > lll"5; ivmiz, 2. 37/vm34; Append.mi Append.l
lo. l4. 2/limi3 ii.23.2/mn27; vin3 i6. 2. 6/nnn8*. 24 4, 31; viu>5> 14; viryo 2. 38/vm46 Astypalea/vmni2, 15
I0.20.6/VD44 ll. 24. l/vnn2l, 29 l6. 2.7/imi 50. i5. 4. 2/m49 3. 39/VU3 Tralles/ivn2i
lo. 2o. lo/vn46 II.24.6.6/mn2I 16. 2. 8/imni, 8* 50. 15. 8. 7/inig* 4. 20/vm34
10. 25. 1/TIU33 II.27. I/IIU22 l6. 2. lo/imi 4. 38/ryni5 PAPYRI:
II. I.4/VU36 ll. 28.2/ivn63 l6. 2. i4/imi INTERPRETATIO CTH 5. 24/VU3 P. Beatty Panop.\
ll. l. 7/ivn69; vnn30, 55 il. 28. i3/umii*; ivn63 i6.2. i5/im2; vmi5 5. I7-I/VIU125, 5° 5. 35/vm46 i/imio*; iimnx-3*, 10
ll. l. l2/ivn58; vn5 ll. 30. l2/im9 i6. 2. 26/vn56 I2. I.2/vni4 7.45/Ivrl62 2/mnn2*, 8*, io; wn2i*
ll. l. l4/ivn70; vai4 l2. l. l/imn8, 22; vm36 i6. 8. 3/mi22 8.23/vm46
ll. l. l5/nm34 I2. I. 3/IIU22 JUSTINIAN P. Cair. Isidor.
II.I. I9/VU54* I2. I.6/VU26 Nov. CHLOTH. i/iimn3, 8, io; rrnn22, 27;
ll. l. 26/vnni4, 65 I2. I. IO/IIU22;IVD48;VH26 DIGESTA l7. 8/ivn6l; vmi2 Edictum 8/vnil VU9
11. 1.28/imii l2. i. ii/ivn48; vn26 2. i4. 42/l"tul9, <5l l28. l/mn8*; vmi2 8/IVU22*
li. i. 33/nm2i I2. I. I2/UU22; VUI4 7. I.7. 2/B127 l68/vmi2 COD. DIPL. LANGOB. 42, 102, 103/ivnii*
XI.2.4/VU54* i2. i. i3/ivn48; VU26 7.27.3/m27 Sanct. prag. ad petit. Vigil.) vnnii, 22
ii. 3. i/rvn6i; vn6o; vm38 I2. I. I4/VU26 i9. l. 52/m3l l8/vin46 P. Ital.i
ll. 3.2/ivnn3l, 58; vnn4i, 67; l2. l. l7/vm36 l9-2. l9.4-5/ivn4* 26/VIU46;Append.in3 DE FISCO BARCINON. I/IVU26
vmn3 8-9 I2. I. I8/VU26 I9.2. 32/VU37 vnii 3/vm4
ll. 3.3/vm38 i2. l. l9/mn3l; rvn5l 25. 1. 13/11127 NANTH
II.3. 5/IUQ2I; ivn6l I2. I.22/IVU48 28. 5. 3. l/vni7 3/IVU3 EPIST. CONSTANT. AD AR. P. Mich.
ll. 4. l/vm40 I2. I.23-7/IU72* 30.l. ll2.pr./vn5* ivnn38, 4i 620/Append. m28
 150 INDEX OF SOURCES
INDEX OF SOURCES 151
P. Oxy. i905f/nm8; nrnnil, CASSIUS DIO V. Canon.
SIDONIUS 31. 2/11138 4. 7l. 3-4/m38
26* 52.28.3-4/vm27 3/umi4* Carm. 32. 9/m39 5. 5. 2/m32
53.22.5/IU52 l3/ivn26
62. 2. I/VIU28 TANEGYRIQUES LATINS
Ann. 5.25.4/U14I
SYBO-ROMAN LAWBOOK
i. li.4/mn4, 19
Legessaeculares 62. 3/mni9*, 25, 36 8(j). 5. l/ivni2 SUETONIUS 1. 31/11152 TERTULLIAN
ii7/im8 7l. 3-3/m68; VIU3I 8(5). 5. 5/ivn27 Apo\og.
Vespas. 2. 5/IU29
i2i/mn6; miip 8(5). 6. i/ivnni3, 27 l6/mni9, 42 2. 6/IU52 l3. 5-6/vm20
COLUMELLA 8(5). io. 5/ivni2 2.42. 5/M26* 42.9/mi28; VM22
TESTAM. S. CAESAR. ARBL. De re rust. 8(5). ll/ivnni2, 2$, 29 SYMMACHUS 2.47.2/11130
ivn26 i. 7/vn8 8(5). i2. 3/ivn28 Epist. 2. 5<S/m29 THEODORET OF CYRRHUS
8(5). l3. l/ivni2 5. 87/VU32* 2.60.4/IU4I Epist.
URSO. CHARTER OF EUSEBIUS OF CABS. 8(5). l4. 3/ivn33 7.66/VU32* 4.6.4/m43 42/vmnio, 29*
vm4 Hist. ecd. 7. l26/ivn62; vn32* 4. 72/mn29, 37 47/vmni4, 29*
PLINY
I0.7/nn7 9. 6/vn32* 6.4i. l/m35
LITERARY V. Const. Epist. 9. io/vn32* i2. 58.2/m30
VICTOR, S. AURELIUS

3. 19/vmiS De Caes.
4.2-3/vmi4 9. 40/VU32* l2. 62-3/mn30, 40
4. 14/vmiS 39. 3i-2/ivni8; vm23
Acta Maximiliani 9. i40/vn32* I3. 50/IDQ5, 19
7. i8/vmni8-l9, 21 Relat.
i/imi8 EUTROPIUS I3. 5I. 2/IQ34; IVB4I VITA CAESAR. AREL.
9. 15/vmiS
Breviar. 3. 15, l8/vn3 14.46/11152 I.20/IVI126
9.36-7/vxniS Hist.
AMMIAN. 9.23/IvnI8
io. i7/m6i
Rer. gest.
TACITUS l. 46. 3/mn37, 42 VITA MEIANIAE
lo. 37/l°6l Agricola
3I. I4.2/1VU54 GREGOR. TURON. 2. 94. 6/IU33 20/IIU24
10.39/IoiSl l3. l/m38
3I.4.4/IV054 Hist. 4.26. I/IU40
lo. iio-ii/mfil
3. I5/VU25 l9. 4/I"38 4. 32. 5/I"36
io. i 13/m?2
APPIAN 4. 2/ivn23
Bella civilia 9. 30/rvnn23, 26* Justinian. Sanct. pragm., in Corpus luris Civilis, m, 799-802
PROCOPIUS
5.4/l"59 Wars (de bell. Goth.)
P. Beatty Panop.1-2, inT.C. Skeat,cd., PapyrifromPanopolisin theChesterBeatty LibraryDublin
JOHN LYDUS (Dublin, 1964)
5(l). 3.2-4/vm46
APULEIUS De magist. P. Ital., in J.-O. Tjader, Die nichtliterarischen lateinische Papyri Italims aus der Zeit 445-700, i
5(l). 3. 9/vm46 (Lund, 1955)
Apologia l. 4/ivmi9-lo, 18-19
7(3). 6. 5-7/vn32* P. Oxy. 1905, analysedin Deleage,Capitation, pp. 73-5
loi/m48 3.47/IVU7I* 7(3). l3-i/vn32* Basil, St., Epist., in Saint Basile, Lettres, ed. and tr. Y. Courtonne, 3 vols.. Collection des Uni-
7(3). 22. 2&-2I/Vn32*
JOSHUA STYL. versites de France (Paris, 1957-66)
AUGUSTINB
De civ. dei Chron. SALVIAN

lo. l.2/vn46 83-4/VU32 De gub. dei

Enarr. in Psalm. 5. l7-l8/m6i*
ix8.8. l9. l/ivn4 LACTANTIUS 5.42-3/limiS
Epist. De mart. pers. 6. l8/vm43
88/nn6 7/ivnn9-ia
23/m54> ivmiio, 23-4, 31, S. H. A.

AURELIUS. see Victor 37. 47, 65 Antonin. Plus

26/ivni3 7. 8/mi9
BASIL, ST. 3fi/ivni4
Eput.f SENECA

l04/ivn66 LIBBR PONTIFICALIS De benef.

285/WU26* V. Johan. v 7. 5-2/ivnS
2/mni4*
 INDEX OF MODERN AUTHORS 153
Institutions imn20, 27 Lot, F., Etendueet valeur" mn7*; ivnn26*,
- "Lcgis. relig. " imni, 3, 8* 29*; Appendix mi
Gibbon, B., Decline andFall Introduction ni - Impotfancier Introduction nn3-4; mnn23,
Goffart, W., "From Rom. Taxat. to Med. 26: ivn20

Seign. " imz; limn7*, 14; ryn29*; vnn22, - "Lejugum" ivn29*

II SHORT TITLE INDEX OF MODERN AUTHORS
25; vm34; Appendixm29 - Nouvellesrecherches Introduction113
- "Venal Siiffragium" um25 Maitland, F.W., Domesday Bookand Beyond
- "Zosimus" im28;vin2 vn4
Grelle, F., Stipendium mn23-4, 31, 44, 51, 71; Marquardt,J., Staatsverwaltung 1111114-15, 17,
nnn8*,21; mn37;ivnp* 53;vni3
Hahn, I., Theod. Cyrus" vin29 Marrou,H.-I., ReviewofMazzarino,
Hardy, E. R., Large Estates vn22 Aspettisocial! vn54
Heitland, W. E., Ayicola vnny-S, 60 Mazzarino,S., Aspettisociali 71154
Hodgett, G. A.J., Soc. andEconom. Hist. Mickwitz, G., Geldu. Wirtschaft mn};-va;4
vm6 Millar, P., CassiusDio vmzj
Full references are given in the first citation of each work or article. Notes marked with an Huizinga.J., Homeludens m64 - RomanEmpire in;9
asterisk are to be found in Appendix n (beginning on page 122). Immink, P.W.A., "Prop. ou seign. " vn3 Mommsen, Th., Athen. Stiftungsurk."
Johnson, A. C., "Domitianus" mnio* Appendix in4
Abbott, F. F-, and A. C. Johnson, Munic. Cantor, N., Perspectives vrifo* Johnson,A.C.,andL.C.West, Byzantine - Staatsrecht inn8-io, 17, 20-1
Admin. mn6, 57-8, 6o, 72*; nnni9-20; Cerati, A., Carac. annon. Introduction 115; Egypt mn3;mi9;vn22 Monier, R., et. al.. Hist. desinstitutions Intro-
nm7; mi45 mni9, 23;mnn4*, 15, 19, 35, 37; rvnn26- Jones,A.H.M., "Aerarium" inyo ductionn4
Admolfi, M., "Iscrizione inedita" ivn9* 7*, 30*, 67*;vn54; vmn8, 12, 15; Appen- - Ancient Economic History vn50 Moreau,J., Sourceschretiennes39 ivnnio,
Andreades, A., "Deux livres recents" dixmnj, 14, 29 Capitatio Introduction 05; inip; umni6, 14-15, 17
Appendix 29 Chastagnol, A., Review ofCerati, Carac. 23, 36; ivnn26, 29, 45> 72;vn66 Nesselhauf,H., Patrimonium" mjo
- Greek Public Financevm2l annon. Introduction 115 - CitiesofEast.Rom. Emp. m6z Oertel, F., EconomicLife" mjz; imzfi*
Ardant, G., Hist. de1'impSt m2;nn26*; Cipolla, C., Economic Decline vmi - Colonate" mi4 ; vnni, 2, 13, 61 - "Niedergang" m62
vmn5, 7, 1?, 26, 41 Clausing, R., Colonate ivn4; vnni, 5*, 13 - Greek City mni2, 24, 56, 64; imn20, 26-7; Ostrogorsky, G., "Steuersystem" nini4*
Barnes. T. D., Tertullian vin20 Corwin, E. S., ed., Coiu(. o/'u. s. m3 llin2; vmig Pallasse,M., "Capitation" ivn45
Beaudouin, E., Grands domaines vn26 Cullen,C.E.,"Rom.RevenueSystem" m7 LRE Introductionn 5; unn3, 8*, 13; - OrientetOccident mn23
Berchem,D. van, "Annonemilit. " nm2; Delcage, A., Capitation Introduction mi2-4; mna2, 8*; ivnn4*, 9, 20, 29;vnn32, 50*, - "TablettesAlbertini" vn69
ivnio limry, 7-8, IQ*-II*, 19, 36*; rvnn9*, 22*, 54, 56, 59;vmn3*, 29*;Appendixmi6 Perdval, J., Seigneurial Aspects" vm4
- Armee deDiocletien imi4*;ivn42* 38 ; VU56; Appendbs:mn2, 4-5, 7, 9, ii, - Records" Appendbcmn5-io, 13-14, 16, Pharr,C., Theod.Code vnni2, 19
Blake, R., Disraeli mz 15, 17-22, 25, 29 24-7 Piganiol,A., "Capitation" rynn20, 63
Bloch, M., "DependentCultivation" vnn28, Demougeot, E., Unite a la division vni3 Jones, C. P., Review ofSherwin-White, - Emp. ckret. nn24*; lim8*
32* Dolger, F., Byz. Finamvenv. iimi4* Letters ofPliny 11172* - Hist. de Rome 111117*
- Melanges historiques mni9;rm7i* Domes, H., SeVostzeugms ivn38 Karayannopoulos,J., Finamwesen Introduc- - ImpSt decapitation ivnnio, 44
Boak,A.E.R.,andH.C. Youtie,Arch.of Dupont,C., "Privileges' imni, 3, 8* tionn5; mn24; rvn35; Appendixin29 Impot desclarissimes 1111130
Awel. Isid. mnio;ivnn9, 18, 22* Ensslin, W., "Reforms ofDiocletian" nmy - Die Theorie Piganiols" Introduction 115; Polanyi,K., "Aristotle" m65
Bott, H., Grundzuge limip; ivnio Faure,E.,"Etudedelacapitation" ivnn22*, imi6; mnn7*, 10*, 14*, i6; ivnn22*, 31, - Prim. Econom. m2
Bowersock, G., Creek World m-ji* 30*, 45 44, 66;vni Remondon, R., Crise deI'Emp. rom. vni3
Bowie, E.L., "Greeks and their Past" m66 - Italiaannonaria" mi30* Kornemarm, E., "Problem des Untergangs" Sagnac, Ph., Fin deI'Ancien regime vm5
Brockmeyer,N., Arbeitsorganisation vn28* Ferguson,W.S., Hell. Athens mi26 IIU2I Saumagne, Ch., Da role deVorigo" vn46
Broughton,T.R. S., Rom. AsiaMinor Finley,M.I., "Aristotle" 11165 Laet, S.J. de, Portorium m6 Scheltema, H.J., An denWurzeln vnfip;
Appendixm24 Frank, R.I., Ammianus on RomanTaxa- League,R.W., Rom. Private Law vni6 vm29
Brown, P. R. L., "Holy Man" vn32* tion" iim2*; ivnp* Lemerle, P., Esquisse" vn69 Schwahn, W., "Tributum" mnil-i3, 16-
- Review ofjones, LRE: Introduction n2 FusteldeCoulanges,N.D., Alleu ivn4* - ReviewofDeleage,Capitation ivnn20,45 i7, 26, 44 5><5o;Appendbc in2*
Buckland, W. W., Text-book ivni;vmi7, 16, - "Colonat romain" vni Lepper,P.A., ReviewofSherwin-White, Schwanz, E., Gesamm. Schrift. ivn38
37 Gaudemet, J., "Civilis" nni7 Letters ofPtiny mjz* Seeck, 0., Regesten Introduction nz; rvn6l;
CalluJ. -P., Polit. monet. nmn2*, 5-6; - Constantm et les curies" 11024*; rvn48; Liebeschiitz,W.,"MoneyEconomy" vryo
ivnnio, 22* vni4 VU54*;vm46 - "Schatzungsordnung"vn59;Appendbc
 154 INDEX OF MODERN AUTHORS

Stevens,C. E., "Agric.andRuralLife"

Segre, A., Colonate" ivn4; vn22 mi 9; mn7*
Studies" mnn3, 7, ig, 36*; rvni5; iustaj., "Zur Geschichte" Appendixinz*
Appendix in2* Svoronos, N. G., "Recherches sur Ie cadastre
Seston. W., Diocletien mm20, 44 byzantin" vm28; Appendixin29
Sherwin-White,A.N.,Citizenship m62
Ill GENERAL INDEX AND GLOSSARY
TeallJ., Age ofConstantine" mi; ivn33
- Letters ofPliny 11172* Veyne, P., Institution alimentaire" viruy
Sickle, }, van, "Diocletian" 111121 Visscher, F. de, £A'fa im26
Skeat, T. C., Pap.from Panop. iinio* White, K.D., Roman Farming vn32
Stark, W., Contained Economy vm32 Wilamowitz-Moellendorf.U. von. and
Stein, E., Bas-Empire vn28 B. Niese, Staatu. Gesellschaft 11026
Zur Gesch. der byz. Finanzw. " Iimi4*; Wilson, Ch., "Gov. Policy" vn59
Taxat. andDecline vmi

abstraction, fiscal 32, 34, ySn, 96, lo6, 108, transport) 35n, 48n, l3on. See also
i26n. See also caput^ geikon, iugum, man- animals, munus
sus, millena, money, stichos, zugokephale animals: assessment 37, 45, 5in, 52, 6o, 63,
abuse: by officials or taxpayers 11, 13, 14, 64, 95, 96, 97> I06> II4-2I piissim, 13 in,
iyn, 32, 34, 107; social i28n, 145. See I33n; tax on lin; acquisition 109. See
also cwiales, fairness, tax evasion also horses, schedule
acceptio personarum (inequality before law) annona (tax-in-kind): institution 6, 31-4,
107 i26n; significance 67-8; other contribu-
Achaeia i22n tions in kind 14, 105, l24n; as rations 31,
adaeratio (commutation of taxes) 83, 84, ioi, i3on; mentioned 411, 36-54 passim,
iiin, i26n, i3on, i36n. See also translatio 8l, 84, 94-5, IOI, 106, llin, 12711, 1360.
adcrescentes:defined 59; mentioned 26, 58 See also adaeratio, annonocaput, praestatio,
Adrianople, battle of (378) 13611 tax

adscripticius (type of colonus) 77n, Spn, I34n. annomcaput (assessment unit in Byzantine
See also censibus adscriptus Italy) ll6n, l28n
Aelian (author on tactics) 139, 140, 145 Antiquity, Classical: and taxation 29-30,
aerarium. See financial administration 55> 9I-3> !45; and social welfare 19;
Africa: saltus in i2n; laws regarding 24, money in, compared to modern 10511;
62n; inscriptions from 92; painter from and accounting 14311. See also finance
l36n; mentioned l22n. See also Anullinus, Antony, Mark 17
Caecilian, centuria Anullinus, proconsul of Africa 230
agri desert! (deserted fields): in legal sense 28, Apamea. 12
35n, 6in, 67n, i37n; literally 51, 8o. See appeal, legal 21, 32, 88, 8911, 94, 103, 1230
also cultivation, demography Appian 17
agriculture. Seeeconomy, land apportionment: of public charges 31, 32,
agronomists, Roman y6n, 135n. See also 6i, 94, 96, loyn, i23n; ofannona by
Columella mga 34, 94
Alexandria, church of 3711 Arabia 1290
Amida 135n Arcadius Charisius (jurist) 45
Aiumianus Marcellinus 142, 143. See also Archelaus, king ofCappadocia13
Index of Sources Aristius Optatus, prefect of Egypt 34, l28n,
Atiastasius, emperor 13311 i3on. See also Edict
Anatolius, Praetorian Prefect of Illyricum aristocracy: local l8n, ipn, 1230; official
143" 86n. See also potentior
angariae (supply of labour or animals for Arius (heresiarch) 53n, 1320
 156 GENERAL INDEX AND GLOSSARY GENERAL INDEX AND GLOSSARY 157
army. See soldiers and veterans iugum, kapnikon, kephale, law, "Maurice," casarius. See cottager, inquiUnus clergy (cferici): exemption from burdens
aroura (Egyptian unit ofsurface) 340, i27n, Nicephorus,paroikos, stichos, taxation censibus adscriptus, adfixus (enrolled in public 22-4, 27, 29, I23n> I32n> 1360; denial of
l28n, l30n, l32n record): citizen 58n; slave 61, 63, 79, enrollment among 29, l24n; French
Arrian (author on tactics) 139, 140, 145 cadaster 8, iiyn, 120-1. Seealsocensus,survey l36n. See also adscripticius IQjIl
Asclepiodotus (author on tactics) 14011 CaecilianofCarthage, bishop23 censitor (assessor): role described 52; men- Codes ofTheodosius n andjustinian:
Asia Minor: cities of 17; inscriptions from Caesar,Julius 7, ij tioned 45, 56n, 63n, l29n, 143 languageof40n, 53, 7in; mentioned 3,
53, 6l, 113-21, i43n Caligula, emperor i6n, 200 census', registry of persons and property for 35, 36. 90", 9I. 95> II2i I44- See also
assessment: devolution from persons to land Campania 62n public purposes: provincial 7-9, l5-l6, Digest, law, and Index of Sources
5, 35, 39. 6in, 64-7, ?6-7, »3. 84, 90-1, capital: commercial, of veterans 25; value 20; city as focus 12-15, l3on; unpopu- cohortalis (type of civil servant) 25
95, 99, l°4, III, 144; by city council 8, 37", 52; oflandlords 73, 92; Byzantine larity i6n; of persons on estates 3611, 71, collatio. See conlatio
9, 32, 94; distinct from taxes 11, i2n, tax on i04n; book and market value of 8l, 84, 8511, 88-9, 120; ofGalerius 44-6, collection, of taxes: administration 15, 17,
4?n, 53-4, 55n, l27n, 144-5; proportional 10511. See also money 48-9, 52, 56, 62, 68, 13311; ofCon- 31, 94i 98, 106, ill, 112, l24n, i26n; by
39-40, 95, l26n; revisions 48n, 63n, 88-9, capitalis illatio. See head tax stantiusI 45, 48, 5i; persons exempted curiales 28, 45, I3?n; from tenant-farmers
102, io8-9, izgn, 1430. See also abstrac- capitatio: main discussion 3 5-40; earliest use 48n; object ofioi, 1220, i43n. See also 64, ?2, 74. 76-7, 8o, 82, 83, 86, 87;
tion, census, descriptio, schedule 26n, 47; problem passages35n, 4on, l42n; assessment, conventio, Diocletian, schedule obstacles to 108
assessor. See censitor
as munia agrorum 36, 64; as a tax 36n, - assessment register: Reichcensus 9, 107; colonate, bound: historiography of 420, 66,
Astypalea, inscription from 11411, ii8, ll9n, 14311;plebeia 360, 37-9, izyn; humanaet lands booked in 24-5; objectivity 33, 70, 75, 85-6, 89n, i2gn, l34"> l35n> 13611;
l2l. See also Index of Sources animalium 37, 6on, 61-4, 65n, 66, 8l, 61-2; humanae capitationis 37, 6l-2, 86-8, not in question 68, 70, 72, 74~5; birth of
Athalaric, Ostrogothic king ill 86-7, 95-6, I29a, I34"; of iuga 39-40, 98; includes personal capita 43, 44, 50; 77, 8o-4, 89-90, i36n; mentioned 5, 65,
Athens, inscription from ii4n 43", 5on, 51, 6311, i2pn, 143; relation to cohnus on 77n, 81, 85, 93, 103, 109, iia; 71, Sf-6. See also colonus, fugitive,
Augustus, emperor 7, 8, 9, 13, 14, 15, 20, caput 39-40, 41, 44; as formation of clubs mentioned 38, 54, 58, Sja. See also receiver, tenant-farmer
lo6, 12511 oftaxpayers 50, 141-4; mentioned4, 32, adscriptidus, censibus adscriptus, professh cohnatus, ius 66, 84, 1340
Aurelian, emperor 28, 2pn, 1250 62, 88. See also abstraction, assessment, - tax liability: 61-2, 96n, 105, l28n, l3in. colonus: tenant-farmer 68n, 70-3, 75n, 76n,
Aurelius, M., emperor 20 census, schedule See also capitatio 77-9> II4> I20> I29", l3on; imperial 7on,
Autun: Panegyric in behalf of 400, 44, 45, capitation tax 35, 12711, izgn, 13011, See also centralization. See government I34D; only a legal category 7in, 75;
48, 49-52, 6o, 104-5, I42n; register of head tax centuria (unit of survey) 52, p6n, iz8n, l42n, iuris alien! 71-2; bound tenant 74, 77n,
62, 6311; mentioned 460, 68, 97i 98", capitulariiis: recruiting official 25, 14211; tax 144" 79", 8o-I, 83n, 85, 8p, 95, 98-9, 135n;
lign. See also Index of Sources register 13 in charity: civic 29, 42; royal 480; Christian stability of 7911, 80. See also adscripticius,
capitulum (club oftaxpayers) 142-3 I25U cottager, inqi iilinus, landlord, tenant-
barbarian: settled 42n; recruitment 60; capitus (levy or issue ofhay) 35n Charlemagne, empire of 144 farmer
assessed 13611. See also Athalaric, Boudicca, Cappadocia 12, 13, loin chevay (medieval due) l34n Columella (agronomist) 69, 1350
Cietae, Goths caput: definition 41-4, 6o; unsecured share Chios, inscription from 117, l2l commoner (plehs): rural (rustica) 35, 45,
belongings. Seepeculium of assessment 44-53, 59, 64, 68, 94, 97-8, Cicero 140 47~8, 51-2, 54, 6311, 68, 7in, 84, 88-9,
birthplace. Seeorigo ill, 119, i29n, i32n; heading in census Cietae (barbarian people) 13, l6n l3on; tax liability of 37-9; urban (urbana)
Bithynia 170, 21, 46 48-9, 52, 53-9, 95, I33", 142, 145; equiv- Cinamura 115 45~6, 48-9, 52, 54, 68
Boudicca, British queen 13 alent andaddedto iugum4311,44, 50, cities: and taxation 8, IQ-II, 15, 17, 31, communities. See cities
boundarystone (oros) l3on, 14311 54-6, 62, 83U, 88, 95-6, 98, III, 114", 91-8 passim, 101-2, ill, 112; in the commutation. See adaeratio
bread, baking of 99, 13711 ii6n, 118-20, l3on, i34n, 142; perishable Empire, 18-19, 21, 32, 93, 107, i23n; condicio (legal circumstance): of birth 710;
breviarium imperii (summary of imperial element of assessment 60-3, Sga, 95, izo; structure of finance 27; Uturgical policy various 7711, Sin, 870
resources) 7, 13, io6 unrelated to taxation 42n, 13011, i3in; of Greek 29; retention of taxpayers 58, conlatio (payment): in general 25, 26, 27,
bridge. Seerepairs in Gaul 49, 84, Sgn, 9811, iipn, 13111, 59, 6o, 73, 850; response to iuga and l26n, 14511; lustralis auri 36n, 52n;
Britain: modem 6; Roman iin, 13, 42n, l32n, i42n; relation to colonate 65-6, 77, capita 104-10. See also council, curiales, flebalis 37; nova 12411. See also horses,
l04n 8i, 83, 86-7; in Bgypt and Palestine 84, honor, munificence, munus, patria, politics, money, praestatio, tax
Brown, Peter 141 l3on, l32n; future of the term 53, 65, 84, res publica consilium (consent) 1070
burdens. See apportionment, munus, onus, 88, 90, 94, 95, 97, l33", l34"; economic citizen, Roman: "active" 42, 44, 68, 71, 97, consortes (partners) 75n, T2pn
sarcma, tax
effects 104-10;comparedto column lOj, l33"; loss of to army 59; of "front Constantine I, emperor 23-5, 28, 2gn, 39,
business:tax on i in; private 29, 102, log, capital 14211; mentioned 4, 26n, 35, 39, rank" 142; mentioned l^n, 26, 12311 44-5i 49> 53~4> 61-2, 64, 68, 690, 70,
ll2; "Big" 86n. See also wealth 40, 45i 114, 119, 120 civilis, definition 26n I23U, I25U, I32U
Byzantium 12, i3n, 144. See also annono- Caria 12 in civitas stipendiana 17 Constantinople: city 26n, 95D; church 37n.
caput, capital, Constantinople,geikon, casa 13811 Claudius, emperor l6n See also Byzantium
 158 GENERAL INDEX AND GLOSSARY
GENERAL INDEX AND GLOSSARY 159
ConstantiusI, emperor 45. Seealso census deminutio capitis (changein personal status 47"; land survey in 52;wgumin i28n. of one 99, l27n
Constantiusn, emperor 64 at law) 42, 72 Seealso Alexandria,Aristius Optatus, fasces public! (census) 580
contract: municipal iyn; oftenancy 67, 72, demography ilpn, i32n, i36n caput. Edict fields {ager). See agri deserti, {us, land
Son, 81; breachof68; pactio 78n; com- descriptio (assessment): ofsouls {ammarum) emperor, Roman 10, 14, 18-21 passim, 27, finance, public: role otmunera in 24, 93,
mentum jgn; emphyteusis lopn 6o, 90, 95; terrena 6o; mentioned, 4pn. 49, 52, 55, 93-4, 107, l23n, i26n. See l25n; deceptive appearance in early
conventio (summons) 5yn See also caput, head tax also government, and names of individual Empire 27-8; local 27, 101-2; uniqueness
conversion scale. See schedule detentator. See receiver
emperors of in Antiquity 30, 102; role in economic
convicani. See villager devotio (prompt performance ofpublic duty) emphyteusis. See contract
Corvinius, quaestor 15 history gin; mentioned 3, 5, 7, ip, 20.
64, 75n Empire, Roman: regions 311, 34, p6;
Cos, inscription from 114, iiy See also economy, taxation
Dideiphyta 115 inhabitants 16, 26, 45, ioi; world 27,
cottager 80, 120, 12911. See also inqt iilinus,
financialadministration, imperial: aerarium,
Digest(Justiniancollectionofjurisprudence): icy; end of30, pi, 92; the West 48, 6}n, fiscus, andresprivata 12, 20; officials 12,
tenant-farmer imageoftaxationin 21, 144;mentioned 85n; tax lists 49, 83n; the East 6511, 113; 14, 4°", 47, I27"> I37"; treasury in
council, city {curia):role in taxation 28-9, I in, 15, 42, 45, 72. See also Index of internal life 90; mentioned 16, 103, iop, general 19, ill; privilege in debt collec-
32, 34. 39, lo8, II2; mentioned iin, 23, Sources
i37n> 145, 146. See also army, cities, tion 25, 34n, 76n; accounts payable to
25, l32n, i34n. See dfco cities, curiales Diocletian, emperor: reforms 9, 32-3, 35, finance, government, provinces (Jiscalia) 64, 76. 5ee a/so land, revenue
court, oflaw: u. s. Supreme 6; enforcing 38, 47, 49, 55", 61, 6311, 66, 93, 9<S, IOI, endowment: municipal 10-11,27, 28n;
public contributions 24, 27, 32 fiscal year. See mdktw
l25n, 12711, i3in, i33n, 139, 141, 143, alimentary foundations 103-4; ecclesi- flight. See fugitive
cultivation, agricultural: costs 51, 99; I44"> 145; laws 28n, 52, 54, 6l, 68, I24n, astical i25n
undertaking of 72, 85; when land sold forma censualis (model for declaration of
l33"; alleged census 44, 45n, 4yn, i29n; enslavement. See slave assets) 42
78; legal obligation 67, 8i, 87, 95, 99, retirement 46, 48, 141; mentioned 6, 8, equomm collatio. See horses formula census. See schedule
134". See also economy, land, tenant- ion, 20, 22, 26n, 28n, 93, i42n eviction: grounds for 69; mentioned 70, foundations. See endowment
farmer direct tax. See tax
7in, 86, 134" freedmen 12, 8in, l02n
cultivator (cultor}:ratio offreeto slave 11711; disaster: occasion for tribute relief 12, 14; at exactio plebis. See capitatio Frisians 12, 13
mentioned 51, 8o, Sin, 82, 87. See also Autun 5 in
colonus
excusatio. See exemption frwnentum (levy of grain) 13. See also aiinona
discriptio. Seedescriptio exemption, from public burdens: laws on fugitive: not to be supplied as recruit 58;
cura (official obligation) 25 doctor (medicus):exemption 23, 27; public 4, 22, 6o, 12311; temporary 12, 107; tenant-farmer as 64, 70-7 passim, 79, 8o,
curatores: municipal i-jn, 21; private 38, 3pn service 26n
principles of 2411; vacatio 240 25, 530, 82, 87; slave 82. See also receiver
curial seryice 21, 23, 25, i23n domains. See land
55n, i24n; excusatio 300, i3on; tax flinctio (personal performance): contrasted to
curiales (city councillors): restrictions on Domesday Book 1430 immunity 3yn, 38, 39, 46, 108, l23n payment {praestatio) 2711, 4on, 63, 97,
changing status 30, 56, 74; contributions domimum (ownership, overlordship); pre- l34"; personal 1300. See also clergy, 144; public 28, 8on, 105, 12411; of a field
of 37; wider liability 38, 13211;decurions tended 54n; over coloni 8in; extravagant doctor, soldiers and veterans 35, 36n> lion; of churches 370; of minors
42n, io2n, 119, i37n;origo ofyin; abuses ill. See also landlord
expenditure, public: imperial 7, 31, 103, etc. 38-9; mnua. 82; of Italy 101; men-
ofl23n; mentioned i5n, 23, 28, 29, Spn, dominus (owner, master) 7911, 8on, Sin, lay, I27n; local 93
loyn, i43n. See also cities, council,
tioned 24, 26, 1230. Seealso conlatw,
8zn, 85n, »7n, i35n exploitation, private 29, 8o, 92, i36n. See
munus, nominatio
land tax, munus, praestatio
duumvir. See honor also business
currency. See money dux (military official) 750 extraordinaria. See munus, taxation Galerius, emperor 44-6, 48-9, 52ti, 5611.
custom: public 24; "nouvelle (et mauvaise)" See also census
46n, yon; tenancy by 69. Seealsoportoria, earnings. See peculium fairness, in taxation: standard of in census Gaul: census in 7, 15, l6n, 2on; contribu-
vectigalia economy: ideas about 7n, 19; local 14, l8- 46n, 7on; lack of6i, 88, 95, l28n; tions 12, 13, i22n, 143; small farmers in
20, 5i-2> 102-10;recoveryby 17; history imperial concern for 69; mentioned 24n, 36; laws addressedto 8311, 84, 88-9;
damni (damages) 80, 82 21, 67-8, pl, 99, 112, 136"; conditions 3on, 3211, 34, 8311, 107 vicarius, 88. See also Autun, caput
debt: ofcities 28n; ofcultivators 51, 82, 31, 45> 5I", 84, lio; and colonate 75; family: members: sons ofvarious profes- geikon (Byzantine assessment unit) i26n.
i34n; to fields, 82-3. See also security Dutch 9in; primitive 92-3; agricultural sionals23, 25-6, 28; various members 38, See also abstraction
decurion. See cwiales
99, ioi-2, iio, 139; heterogeneity of 39. 45> 50-1; jilius familias (son in patria generosity. See munificence
defensorcivitatis (municipal official) i5n 107. Seealso capital, money, society potestas) 57 georgos (Lat. colonus) 114. See also tenant-
tleficientes (citizens enrolled but absent) 26 Edict, ofprefect ofEgypt (297) 34, 6l, 141. - generally: of soldiers and veterans 28, farmer
Deleage, Andre 113-21 passirn, l^jn. See See also Index of Sources
53-9; guardianship 38-9; and deminutio germanicus 12, 15
also Index of Modern Authors Egypt: Ptolemaic 6, i8n; Roman policy capitis 42; size 43; census of 45, 50, 51, 52, Godefroy, Jacques 4, 6oa, 13911, 14511
delegatio: imperial tax order 106, i27n; toward 17-18; taxation in 31, l3on; 63n, 64, II5-l6, iip, l3on, I36n; of gods, pagan: temples of 51; priest of 119;
money tax ison. See also indictio alleged census in 4.4x1, l2pn; rebellion in slaves 79; ofcoloni 80-2, 8y, 1360; farms mentioned 6jn, 100
 l60 GENERAL INDEX AND GLOSSARY GENERAL INDEX AND GLOSSARY l6l

gold. See money, tax Illyricum: prefecture 77, 80, Sin, 830, Julian, emperor 38, 4on, 143 vation, custom. Digest, Edict, ius,
Goths 6on 143"; province i22n Julius Sabinus, censitor 1290 tenant-farmer, and individual legal terms
government, Roman: imperial: centraliza- Immunity. See exemption Justinian, emperor 35, Sgn, 113 Lesbos, inscription from 114-15, i2o-i
tion 7n, 9, 8pn, 93, 107-8, ill; central incensitus, defined 59 Licinius, emperor 46, 52
15, l8, 96, 103-4, 107, lo8, ill; remote indictio: imperial tax order 13, 27, 31, 33, K(epha!o) Z(ugon) 116 liturgy" (Lat. munus):person performing
from localities 62, 107; as absentee land- 43, 47-8, 5°, 6311, 92, 109, l26n; fiscal kapnikon (Byzantine tax) 13411 i7, 27; and taxation 2411, 2jn; mentioned
lord 76n; and tributary land 101-2, i22n; year 31 kephale: captit 115, 117-19, l32n;
as in 29, 93> I05> i23n, l25n, 1320
bureaus of 105, lo8; mentioned 10, ll, infimiis (citizen oflow stratum) 30n, 141 Byzantium l26n Livy 1411
13, l4> 17, 19, 20, 6o, 64, 68-9, 76, yyn, inflation. See money Kyrene, l22n lochos (file of soldiers) 140, 145
78-9, 82, 84, 89, 92, 93. 98, 102, 110, inqwlinus (house-tenant): defined l2gn; Lucius Titius. See Titius
ii2, l23n, l34n, l3fin. See also financial mentioned 42, 72n, 7711, 8in, 82, 8gn, labour: public 3on, 48, 92; forced 73; Lycia-Pamphylia46
administration, soldiers and veterans 115, 117, 120. 134" supply 79; price 80, 99; on estates izon Lydia l2in
- local, 10, ii, 15, 20, 75, 93i 101-4, 112, inscriptions. See Africa, Asia Minor, Labour Theory of Value 311, 5 in, 55 Lyons 2011
l22ti, l2}n. See also cities Athens, Tenos, Volcei land: imperial 12, 3pn, yon, 7111, ySn, 82,
governor, provincial (praeses): cannot inspection, tax: agents of 8911, 99; men- 86n, 92, ioi, 103, ill, l34n; municipal Macedonia l22n
exempt curiales 28; superyision of cities tioned 4811, 88-9 28n, 100; qualitative categories 33, 340, magistrate: imperial 14; local and unpaid
32, 94> I03> I07> l26n; conflict of interest institutions: history of late Roman 4, jfn, ll4, 120, l26n;and mgum 33, 83, 98; 15, I?. 25, 26, 27, 105. See also honor
76; mentioned 14, 24, 27, 56n, 62, 77ti, 83, 85, 94; ofPrmcipate 7, i3In; inter- arable 33, 42, log, 113-15, i28n; moun- Magnesia on the Meander, inscription from
i3on, 13211 ference with l23n tain and waste 33, 51, 109; rented or 118-19, 121
yammaticus (elementary teacher) 23 Icmian islands 113 leased 44, 68, y6n, 100-2, 1350; manage- mancipium (slave) 12, ia,n, 370
grapevme 33, 42-3, 45, 109, 114, 115 Italy: tribute in 9, 47, 101; landmeasures ment 7611, 78, icon, 13511; development mansus (medieval unit of assessment) l32n,
Greece: culture l8n; cities 29; ancient 142. ll3, l28n. See also Sanction 8o, 109; abundant supply 8911, l36n; 145"
See also Athens, tactics wgatw:asprofessio 35; as itigum40; lucrative urban 117. See also agri deserti, assessment, master. See dominus, landlord
Grelle, Francesco 144. See also Index of 63n; as iugavel capita64-5; mentioned cultivation, economy, property, survey matiere imposable (taxable assets) 99,
Modern Authors 26n, 62, 87ti, 95, 109, 12911, 13311, 141 land tax: definition 12511; early instance 101-2

lugerum (measure of surface) 33, 43, fi2n, 36n; modern idea 37-8; presuppositions Maurice" (author on tactics) 1400
Hadrian, emperor 139 l26n, l28n, 1370 of 67-8; relation to rent 100-2; men- Maxentius, usurper 46
head. Seecaput, deminutio capitis iugwm: meaning 32-5; shareofassessment tioned 8-11, 14, 79, 88, 93, i22n, 12411, Maximinus Daja, emperor 46
head tax 8, 9, ID, 35, 36, 37, 41, 45, 4711, 39, 5°, 53, 65"> 66, pSn, 114, il7> 142; izyn measure. See survey
55-6, 6o, S^n, 93, 100, l22n, l2jn, i2Sn, secured share of assessment 41, 60, 63, landlord: as taxpayer 26-7, 44, 67, 6p, 76, mediciis. See doctor
l30n, 13211, l33n, l36n Ssn, 86-7, 88, 89, 94. 95. 97. l°4, "I> 8l, 98; absentee 37n, 73, 76, 119, l24n, mediocer (citizen of middle stratum) 3011,
Henry n, Angevin empire of 144 II.S, ll9> I20> l27ni l29n, 1330, l37n; i35"; registry by 42, 51, 114-18 passim, l4l
historiography. See colonate, taxation shareofproceeds 106, 1130; yoke of I29U; small 43, 64, 83-4; poorer than merchant. See negotiator
homo stipendiarius (individual taxpayer) 17 oxen 139, l45n; "front rank" 141-2, 144; tenant 43n; greater 64-5, S3, 121, l2gn; millena (Italian unit of assessment) 113,
honestior (citizen of higher stratum) 141. first appearancein laws26n; andland restrictions on 68-9, 74, 78-9. 86, 107-8, il4n, l28n, 1370
See also potentior survey 33, iz6n, l3on; origin 34-5, 47, il2; right to recover fugitives 70, 74-5, modius (measure of volume and surface)
Honor (civic magistracy) 69, lean (diitimvir) 48n, 139; fractional 43n, 44, 117, ll8; 85; potestas over tenants 71-2, 77n, 87; II3-I4. l38n
i24n, 13311. See also council, curiales, economic effects 104-10; assigned to the ofPalestine 85; ofGaul 88-9; imperial money: contributions in 14, 105, 12411,
munus, taxation crown 105, l37ti; in Egypt and Byzan- loan to 103-4; mentioned 80, lop, 13411. l26n, 1300; currency fluctuations 20,
honoratus (imperial dignitary) 37, 7511 tium i28n; eleutherica, tamiaca, chrysotela See also possessor 31-2, 6gn, 9311, 94, 104; payments in gold
horses: contributions of 12, 23, 37, 43"; of l37n; mentioned 4, 22n. See also assess- hrgitiones. See financial administration and silver 23, 82, 87; as a language 32,
imperial post (yeredus) 35n; escorting of ment, caput, census, iugatio, survey law: imperial: dating of 3n; general on 94, I05-6, il2; siliqua 6311; solidus 82n,
mules and 47, 48n. See also animals ius: agrorum 81, 84, 85, 87; alienum 71-2. taxation 9, 83n, 107-8; interpretation i2n; 13in, 143. See also adaeratio, tax
hospitem redpere (quartering) 3511 See also colonatus, law, security loophole in 26n; on curiales 28; contrast munia (obligations to the state): iniuncta
house-tenant. See inquilinus to politics 30, 32; objectivity 33, ii2; and imperil 13; agrorum 36, 95. See also munus
humilior (citizen of lower stratum) 64, 68, Jews:contributionsto Temple 12; andcity real life 75; Byzantine Farmers' 8911; municipalities. See cities
73, l4l. See also mediocer, teiiuior council 28 criminal 141 munificence: municipal 10, 18, 102, l26n;
Hypaepa, inscription from 43, 114-15, i2l Jones, A.H.M. 19, 66, 88, 114-20 passim. ~ private: of persons 5, yo-l; sense of potlatch spirit ofi8n; as standard practice
See also Index of Modern Authors capitt in 42; lease by 68, 78, i35n. See ipn, 27, 29-30, 92. &e also finance
jdoneus (solvent person) 28, 76n Judaea I in also appeal, codes, contract, court, culti- munus (obligation to the state): includes
 l62 GENERAL INDEX AND GLOSSARY GENERAL INDEX AND GLOSSARY 163

taxation 24, 25n, 27, 29, 3on, l23n, 12511; 76-7; and city council 28 praedium (landedproperty) 3611, 6l, 78, rent: imperial revenue from 12, ioi-2, 104;
patrimoniorum 2411, 3511, p2 n, 12411, i2pn; Palestine 83-5, 1360. See also capiit 88n, 11411, l3on, l34n increase 68-9, 86; in relation to taxes 82,
modern misconceptions about 2jn, 92-3 ; Panegyric. See Autun praescriptio temporis (time limit for lawsuit) 100-2, l3in, 13411, l35n, l36n; men-
shares of 3211; Digest text on 3511; munic- papyri 3n, 3 in, 34n, 4»n, 62n, 73n, 113, 77" tioned 67. See also superexactio
ipal38-9, 48-9, 92-3, 102, i36n; personak l28n, i3on, l32n. See also Index of praestatio (payment) 23, 2411, 26, 53, 54n, repairs, of roads and bridges 18, 1370
4$, l24n, l33n; of rural commoners 47, Sources 97-8» 106. See alsofunctio, translatio Republic, Roman 9, 14
67~S, 75n, 77n; prestige 69; sordidumet paroikos: Byzantine Sgn; as mquiUnus 115, Prefect, Praetorian (chiefcivil official) i5n, resjisd. See res publica
extraordinarium 92, l37n. See afeo angariae, Il6, 117 27, 97", i°7n, l3°n, l3?n, l43n. See also res privata. See financial administration
exemption,functio, honor, nominatio, patria (municipality): as petite patrie 27, Anatolius, Aristius Optatus, Edict res publica (state): univocity of the term 26-
ohsequium, onus 93; synonym for res piiblica 5611, 7in, prefecture, praetorian 105, io6, 107, lljn 7; as city 3on, 7in, 101; distinct from res
Mylasa, inscription from 115 i23n; place of registry 70; mentioned 51 principaUs (front-ranker) 141, 144 Jisci 28n. See also cities, patria
patrimonium, of senators 37n. See also mimtis, Principate 7, 16, l8, i3in retentator. See receiver
navicularii (shippers) 1240 onus privilege. See exemption, financial adminis- revenue, imperial: from land 12; stable 13;
necessitas: expenditure 7, i24n, 12511; patrocinium (patronage) 75n, lion, 1350 (ration, tax evasion in money 270; lessees of 28; acquired by
requirement 39, 76n peasant 4811, 75, 99, i2pn, l32n, l35n, 1360. production, taxing of 67, i04n, 13611 officials 106; mentioned 6-7, TO, 12, 20,
negotiator (merchant) 12, ill See also commoner, tenant-farmer professio censualis (declaration): meaning 43; 93, 103-5
Nero, emperor 12 peculium (personal property): of soldiers and special definitions35; ofproperty 13, i6, rich, the: public obligations 0^29, 3on, 42-
Nicephorus, emperor 1450 civil servants (castrense) 25, 54, 56, 57n; 25, 33. 35, 37", 38, 41-4 passim, 94, 96, 3, 48, 63, QI-2, 145; in city council 32; as
nome (Bgyptian territorial district) 24n of tenant-farmers Son, 82, 87 98, loo-i, 104, i05n, 115, iip, 121, 142; social category 57, 69; abuses by i28n;
nominatio 28, io2n, 12311, i34n. See also penates (home) 7in, 7711, 8411 ofiuga 39-41, 49-50, lijn; forms of 42, mentioned 51. See also fairness, potentior,
taxation pensio. See rent 51, ii5> 120-1; and caplit 41, 43, 52, 62; society
norms censa (assessment coefficient) 60, 62-3, pensitafio (periodic payment) 26n, 34n, 530, cannot be transferred 74; mentioned 3on, Rodbertus, C. See Labour Theory of Value
87 6in, 760 48, 6o, 63, 64, 76, 95, 105, l43n. See ako Rome: capital 13, 16; Praetorian camp in
Numidia ill peraequator. See inspection census 46; Senate 74; church l25n; mentioned
perfectissimus (grade of rank) 23 property, real: as security for taxation 32n, ion, 11, 12, 14, 17, l8, 19, loi, 146. See
obligath praediorum (pledge of lands as pertica (measure of length) 3 3 34, 41, 52, 102, I26n; defence ofland- also Empire
security) ll4n phoros (Lat. tributum) 470 lord s against tenant jjn; size 115, l2l. rus (countryside) 81
obsequium (synonym for munus): of re- phylakes (custodian) 14011 See also census, land, possessio, professio, rusticanus, rusticus. See commoner
cruiting 25; contrasted to pensitatio 26n; pignus. See security wealth
offunctiones 39; local 48n, 9211; military plebs. See capitatio, commoner prototypiae. See soldiers and veterans sacramentum (military oath), 58, 59
58n plethra (measure of surface) 33, l28n provinces, of the Roman Empire: contribu- sale: for tax debt 3411;andtax liability 61,
Oea (Tripoli) 15 politics, municipal: and taxation iin, i8a, tions and taxes from 7-9, 13-14, io6, 78, 86, 88, 89n, 107; of slaves and land
officwm: public duty 1311, 70-1, 73; staff of 26, 28, 3&~2, 61-2, 83n, 94 i22n; no movement of slaves from 61, 62, 79> 108; by tenant 77n, i34n; of
an imperial official i5n poll tax 2511, 35, 3<Sn, 66, 88, l3on 79; uniformity among 85; enrichment of tenanted land 78-9; of imperial land 103;
olive trees. See trees polyptychs (estate records), Carolingian 103. Seealsocensus. Empire, governor, mentioned 320, 360, 5yn, 77
onus (burden): collective 12, 13, 14; com- 75", 92", 143" and names of individual provinces saltus (undeveloped land) i2n, 5in, 92
plex of 14, l22n; ofpatrimonia 24x1, 92n; pontus 60, 63, 87 publicans (tax farmers) 15, i6, 17, ig Sanction, Pragmatic (law of 554 for Italy)
of a.praedium 36n, 88n; of navicularii poor (pauperes), the: tax relief for 29, 450, publiws, definition 2611 ll3
l24n; greater i30n. See also munus 4§n; government protection 69, 145; Pudentilla 15 sarcina (burden), oftax liability 58n, 59, 82n
ordo (college of curiaks} 28. See also council sustenance l25n; abused l28n. See also Sardinians 12
originarius: meaning ypn; mentioned yin, rich, society querimonia. See appeal schedule: converting assets into shares of
gpn portoria (internal customs dues) 7, 103 assessment (formula census) 33-4, 4.1, 43-4,
origo (city of birth registry): of curiales 28; possessio (title to property) 36, 8511, 12711 rank and file 139-45 49-50. 52, 60, 62-3, 96-7, 98n, 116-20
of tenant-farmers 70, yin, 72, 73, 74, possessor (proprietor) 7in, 7911, Sin, 85n, rate. See tax rate passim, i26n, l27n, 1300, i34n, 141-2;
7?n, 7pn, 8in, 83n, 84, 85n, 86n 8jn, 88n receiver {detentator, retentator), of fugitives for adaeratio i26n
ownership. See dominium, dominus, landlord, potentior (citizen of highest stratum) 1511, 70, 74. 7<5> 77, 79-89 passim security: for tax debt 13, 34, 39, 52, 76-7,
possessio 3011 (potior), 64, 141, 146. See also reform: fiscal411; at onset ofPrincipate 16- i25n, l2pn; for residence by tenant 73;
honestior, humilwr, injimus, mediocer, 17; of assessment 34; ofTurgot 92; spirit ius pignoris 770; for a loan 103. See also
painter. See Africa tenuior of i33n. See also Diocletian debt, security
palatini (emperor's staff): privileges of25, potestas. See family, landlord Reims 45 seigneurie, medieval 67n, 750
 164 GENERAL INDEX AND GLOSSARY GENERAL INDEX AND GLOSSARY l6$
senators, Roman: as census officials 15; for- also aroura, boundary stone, cadaster, 71, 73-7, 8o-2, 89, 95, 98, 107-8, I34"> nexus Sin; solum tributanum 110
tunes of 19; taxes of 37, 38, 64, 75; men- centuria, iugerum, modius, pertica, plethra l36n; modern 91; "clubs" of 142-4. See tributum (tribute): defined 11-14; capitis
tioned iifin, 119 susceptio (duty of receiver) 24 also landlord, Titius, tributarius 8-ll passim, 5511, i3in; salt 8-9, 15, i22n;
servitium (service) 7411 symmory (club of taxpayers) 142 temonarius. See soldiers and veterans regium 12, loin; total of 20, lo6, l22n,
servus. Seecensibiisadscriptus.slave Syria i in, i22n, i28n, i2pn, 1440 tenant, ofa house. Seeinquilinus ijin; pledge by reason of34n; of res 36,
share of assessment. See abstraction, caput, Syro-Roman Lawbook 33, 35, 6o, 63, li6, tenant-farmer: income of 43-4, 67-9; two 95; as rent 82, loi; mentioned 3, 4, 7-ii
iugum, schedule ll7, l4i> 142. See also iugum, schedule, categories of 64, ^6, 77; binding of6j, passim. Son, 950, loo, 103, 104, 1220,
Sicily l7n, 13in and Index of Sources 69, 72, 76, 77, 8o-l, 8311, 84, 86-7, 9°, 97, i3in, 1360, l3yn. See also munus, taxa-
Sidonius Apollinaris 49, l42n. See also l34n> l35n; customary 69; law of 332 tion, vectigalia
Index of Sources tactics, Greek authors on 139-41, 144-5 about 70-77, 79-84passim;law of 371 Turcius (name ofan inscription) i i4n
siliqua. See money tax: direct and indirect 4, 6, 7, 9, 10, II, about 77-84; law for Palestine about 84- Turgot, A. R.J. 92
slave (servus): as element of wealth 41-2, 14, 15. 18, ip, 20, 29, 3I> 93> I02> I°S. 6; law for Thrace about 86-8, gSn. See
45, 5in, 52, 55, 6o, 63-4; entered in ill, 112; "single" 6; total proceeds 10, also colonate, colonus, contract, fugitive, vacatio. See exemption
census 26, 6l, 114-17, 119-20; enslave- II, 13, 14, I?' I02> IC16; old and new 13; receiver, taxpayer Valens, emperor 44, 53, 58, 6on, 63, 77
ment 42, 7in, 73-4; sale of registered 62, local 27; in gold 33n, 3611, 37n, 38, 84, Tenos, inscription from 10-11 Valentinian I, emperor 39
79, 86, SQn, 108; chaining of 70-1, 73-4; l29n, i36n; on special groups 37n. See teniiior (citizen ofeconomically weaker Valentinian m, emperor ill, 11411
kinship of colonus to 71-4, 77n, 79, 87, also adaeratio, annona, head tax, land, stratum) 39, 141, 142, 144, 146 vectigaks (imperial revenue officials) 25
8pn; productivity ofl35n; mentioned land tax, money, publican, rent, vectigalia, tenure (plot): of slaves 6in; of a peasant vectigalia^. indirect taxes 7, 9, 15, 19, icon;
8in, 109 vestis i2gn land tax 36
society: welfare 18-19, ill; role ofmtinera tax base 4n, 67, 97, 143 territory: of a city 15, 129, l44n; as tax veredus. See horses
in ancient 29-30, 92-3; mobility in 52; tax evasion 29, 30n, 54, 57, 62, 72, 760, district 95, 97, io8, 121; as homogeneous Vespasian, emperor 13
affected by tax law 66-8, 79, 90, ii2; 79", 88-9, 92, 102, lion, l3on, l37n- See state 107-8 vestis (late Roman tax) 36n, 38
concord 92, 145-6. See also economy, also abuse, exemption, fairness, inspection tetrarchy (imperial college) 90, 141 veteran. See soldiers and veterans
potentior tax level 104, l24n Themistius 12711 village (yicus) 34n, ?in, 7811, 103, ll5
soldiers and veterans: recruitment 18, 25, tax policy 17, 20 Theodosian Code. See Codes ofTheodosius villager 73, 75, I30n
27. 37", 38, 57, 58-9, I29"; pay 20, 31, tax process 10, 17, 3 in, 34, 93, 107, ll^n ii andjustinian vine. See grapevine
94, loi; privileges 24, 25, 35"i44. 53-60, tax rate l22n, l27n TheodosiusI, emperor 85 Vitellius, emperor X2
I24n, 12511, l33n; prototypiae 25, 38; tax scheme 11, 95, 97, l3in, 139 Theophanes 14511 Volcei, inscription from 113-14
service compared to civilian 26; discharge tax system 22-3, 27, 63, lio, 112, l28n Thera, inscription from 114, 115-17, i2o volumen (roll) 24
53-4; comitatenses, ripenses, andprotectores taxation: historiography 3-4, 6, y, 8, 9i ion, Thessalonica, church of 3 yn
54, 59"; social origins 56; mentioned 12, 22, 28, 32, 36n, 4°> 45n, 47n> 5°n> 52n, Thrace 12, 6o, 63, 66, 83, 86, 87 wealth: export of 1411, 102; individual ipn,
13. See also barbarian, family, rank and 54n> 55, 6on, 66, 9I-5, 99, 106-7, III-I2, Tiberius, emperor 13 20, 60, 100, 102-3, 105, io8; qualification
file ll4n, l2on, l25n, l26n, izyn, 1310, Titius, Lucius (fictitious taxpayer) 33-4, 43 for public obligations 29; assessment of
solidus. See money l32n, l37n, 139; early imperial 4, 7, 8, 9, titulus (heading in a record): annonarius 53, 32, 41, 5In> 58> 93> i27n; advertised in
sors (allotment) 40 II, l6-l7, 67, 93> 102-4, ill; compara- 54n; cobnorum 81; of taxation 13in census 42, i2i; disparities of 57, 6o; from
Spain l22n tive 6, 7, 27-8, 2gn, 3on, 55> 9:t~~3i 99. Trajan, emperor 139 agriculture 99; new 102; of localities
Spalato, Diocletian's palace at 14111 103, l22n; ordinary and extraordinary Tralles, inscription from 119, I20-I, 1310 lo8-9; of church 12511. See also rich,
stichos, defined 145 13-14, 107, 110, III, I24", 132"; ofin- translatio (transfer): of pmestatio 150, 84, society
stipendium: tribute 14, loan, i22n; years of dividuals 1411, i5-l6, 20, 94, 107-10, II2, l36n, l3yn; ofprofessio 74
military service 53, 5pn l27n; profit from managing 17, 12311; trees: olive 33, 42, log, 114-15; assessment zugokephale (share of assessment) 87, 95, 96,
stipes (voluntary contributions) l2n relation to nominatio. functio, munera, and of 45, 6an, 630 9§, ll6n, 118, l34n, 142
stoichein (to be in line) 140-1, 145. See also honores 23-32 passim, 39, 92-3, 102, 105, tribune (imperial dignitary) 119 zugon: as Lit. iugMm 98, ll6n, 117, ii8,
stichos i23n, 12411, i25n, l3on; ethics of 29-30, trihHtariiis:as non-citizentaxpayer 4211, 82; ll9, l45"; as front rank" 139-41
suffragium (solicitation) 38 92; of minors etc. 39; abatement 45-6,
superexactio (collection of rent or tax above 50, 58, 62I1, 6711, 86-7, 97> 98, I22n;
fixed level) 68-9, 13511 Byzantine tract on 145
survey, of land: for tax purposes 8, 45, 47, taxpayer: early imperial 10; relation to
i22n, 13111; irrelevant to iugum 3311; units government 20-1, 31, 6o, 91-2, 104-10;
of 33. 34> 95. T.T.3-1J passim; needed for widening circle 28-9, 39, 47-8, 52, 67,
professio 44, 48, 52, 64; surveyor 144. See 89, 94*> poorer 64; tenant-farmer as 67-