LEWIS, NAPHTALI, A Reversal of a Tax Policy in Roman Egypt , Greek, Roman and Byzantine 
Studies, 34:1 (1993:Spring) p.101 

A Reversal of a Tax Policy 
in Roman Egypt 

N aphtali Lewis 

KRIEF NOTATION in a lengthy, recently-published papyrus 
roll, when juxtaposed with data available for half a century 

and more, documents an unprecedented reversal of 
governmental tax-collection policy from collective to individual 
responsibility. 

1. The Old Evidence 

In non-nomadic societies the basic human impulse is to stay 
put; but when conditions become intolerable, flight remains a 
last resort. The advent of Ptolemaic rule in Egypt imposed, 
among other things, a monetized economy upon a countryside 
still rooted in barter. When that fundamental contradiction 
presented them with situations beyond their ability to under­
stand or cope, Egyptian peasants and workmen would down 
tools and take to their heels. 1 Usually they would make for a 
nearby temple, where they could find asylum while awaiting a 
resolution of the difficulty. 

That counsel of despair, flight, continued to play its role under 
Roman rule, but with a significant difference. Under the 
Ptolemies the flight from the job was in the nature of a strike, 
with the workmen waiting to negotiate a return to work; in the 
Roman period the flight was often a final solution: the fugitives 
would leave home with no hope or expectation of an early, or 
even eventual, return. This radical change resulted principally 
from the pressures of taxes and liturgies, which were applied 
more ineluctably under the Roman than under the feebler 
regime that preceded the Roman annexation of Egypt. 

Unauthorized removal from one's idia, or registered place of 
abode, resulted automatically in outlawry. Such flight was 

1 For a case study see the brilliant analysis of PSI V 512 by J. Bingen, 
Proc.Congr. XII (=Am.Stud.Pap. 12 [1970J) 35-40. 

101 



LEWIS, NAPHTALI, A Reversal of a Tax Policy in Roman Egypt , Greek, Roman and Byzantine 
Studies, 34:1 (1993:Spring) p.101 

102 REVERSAL OF A TAX POLICY IN ROMAN EGYPT 

termed avaxwPllCJu;. The image from the underlying verb 
(avaXCJ)pECJ)) is that of "going up" into the hills flanking the Nile 
valley. From those hideouts the fugitives lived by banditry, 
raiding populated places in the countryside for food and other 
loot. A certain number-we have, of course, no statistics­
made their way to the cities, looking to disappear into the 
protective anonymity of the urban melting-pot and eke out at 
least a marginal subsistence there. 2 

Missing persons meant unpaid taxes, anathema to the Roman 
administration but grudgingly and temporarily accepted in 
times of crisis. One such instance occurred in the last years of 
Claudius and the first years of Nero. For reasons not stated in 
the documents,3 the village of Philadelphia in the Arsinoite 
nome (the area of the present-day Fayum southwest of Cairo) 
suffered a loss through avaxwPllCJl<; of more than ten per cent 
of its adult male population. Among the documents relating to 
these developments is one in which a group of poll-tax farmers 
petitions the prefect of Egypt for a reduction of their contrac­
tual obligation because 

from their previously populous condition the aforesaid 
villages have been reduced to small numbers through some 
men having fled leaving no property and others having died 
leaving no next-of-kin, and therefore we are faced with the 
danger of having to abandon our collectorships because of 
lack of resources.4 

2 E.g. Caracalla's edict of 215, P.Giss. 40 ii (Sel.Pap. 215), ordered the 
expulsion of "countryside folk who have fled to Alexandria from other parts." 
Such evidence led A. C. Johnson, Roman Egypt (=1'. Frank, ed., An Econ­
omic Survey of Ancient Rome II [Baltimore 1936]) 354 (similarly vi, 250), 
following the lead of E. Bickermann, Gnomon 3 (1927) 671-75, to the view 
that those flights were merely "part of an urban movement to join in the 
industrial activity of Alexandria, where life was more varied and less 
precarious than in the rural sections." While the city undoubtedly had its 
attraction for some of the country-dwellers, the overwhelming evidence of the 
papyri clearly defines avaxroPll<H<; as essentially resulting from economic ruin 
or distress (cf Fras~r 16 infra). People reduced to that condition were termed 
exasthenesantes, on which see P. Thmouis 1 p.29. 

3 In the absence of evidence all suggestions must remain conjectural. For 
Egypt one naturally thinks first of a series of low Niles producing inadequate 
harvests: so e.g. D. Bonneau, Le Fisc et Ie Nil (Paris 1972) 236. On the 
depopulation of Philadelphia (which started, however, not in 45 but later: cf 
JEA 23 [1937] 65 n.3), see P.Mich. X 594 introd. 

4 P. Graux 2 (SB IV 7462; set. Pap. 281). 



LEWIS, NAPHTALI, A Reversal of a Tax Policy in Roman Egypt , Greek, Roman and Byzantine 
Studies, 34:1 (1993:Spring) p.101 

NAPHTALI LEWIS 103 

The prefect's reply is not preserved, but there is every reason 
to suppose that he granted the abatement as a pis aller. We do 
know that he proclaimed a partial tax forgiveness that induced 
forty-seven of the 152 missing men to return home.s 

Other papyri tell us of two such philanthropa under Hadrian. 
The first, occasioned presumably by the ravages of the Jewish 
revolt of 115-117, was a sizable reduction of the tax in kind to be 
paid by tenants of public and imperial lands; for example, one 
carefully detailed document shows reductions on different par­
cels of some fifty to eighty per cent. In the second century (136) 
the emperor, in view of two successive years of "insufficient 
and incomplete" Nile floods, granted the farmers a tax mora-
torium of two to five years. 6 . 

Concessions like the above were, however, but occasional 
responses to occasional emergencies. Mostly the trend, even 
under "the Good Emperors," was in the other direction: in the 
case of the liturgies and of at least some taxes, the whole com­
munity constituting the local tax unit (e.g. the village) was made 
collectively responsible for the obligations of its members.7 On 
present evidence it was under Trajan that an inchoate liturgic 
system became full blown, and it was also under Trajan that 
there was introduced a new tax, merismos anakechorekoton, 
which annually prorated unpaid taxes of fugitives among the 
taxpayers remaining in the community.8 

In any given year the amount to be raised by the merismos 
anakechorekoton would vary, of course, with the number of 
delinquent fugitives. From time to time, notably in his edict 
ordering the taking of the fourteen-year census, the prefect of 
Egypt would proclaim an amnesty in order to induce the 
fugitives to return home. 9 Obviously many availed themselves 

5 P.Ryl. IV 595.133-85. On this and related documents see J. F. Oates, 
Am.Stud.Pap. 1 (1966) 87-95, with my supplementary note, BASP 4 (1967) 17ff. 

6 P.Brem. 36; P.Oslo 78 (SB III 6944; FIRA 181). 
7 On collective responsibility for liturgies see my The Compulsory Public 

Services oj Roman Egypt (=Pap.Flor. XI [Florence1982]) 70, 79. 
8 Details on dates are given two paragraphs below. 
9 A good example of such an edict is P.Lond. III 904 (SeI.Pap. 220; W.Chr. 

202), of 90 (amnesty, though not pronounced expressis verbis, is surely 
implied); see also BGV 1159 (W.Chr. 408). BGV II 372 (W.Chr. 19), a similar 
edict, was issued by the prefect M. Sempronius Liberalis not in a census year 
but in order to induce fugitives to return home after the dislocations caused by 
the uprising of 153/154. P.Ryl. II 78 (157) refers to Liberalis' edict 1tEp[l 't ]rov 
avuKEXffiPTl1CO'tffi(V), P.Fay. 24 (158) to his order m:pl 'troy £1tl~EVffiV ... W(J'tE 



LEWIS, NAPHTALI, A Reversal of a Tax Policy in Roman Egypt , Greek, Roman and Byzantine 
Studies, 34:1 (1993:Spring) p.101 

104 REVERSAL OF A TAX POLICY IN ROMAN EGYPT 

of that opportunity to return home, and this fact is reflected in 
the extant receipts for the payment of the merismos ana­
kechorekoton: in the year after the census proclamation the 
payment made by one of the remaining taxpayers was less than a 
tenth of what it had been the year before.10 

The earliest attestation of the tax dates from 103 (PSI IX 1043), 
the latest from 161 CWO 651).11 But the significance of those 
dates has remained unclear: do they reflect a historical reality or 
merely the accident of the finds? With respect to 161 that ques­
tion can now be answered. 

II. The New Evidence 

Papyri are not normally preserved in the damp soil of the Nile 
delta. Exceptionally, Greek papyri of the second half of the 
second and the first half of the third centuries, preserved 
because they were carbonized in the destruction of their reposi­
tories, have been found at two sites, mostly by clandestine dig­
gers-which means, of course, that the finds were sold to 
various buyers through the antiquities markets. Fragments, 
more than 5,000 in number, from some three dozen rolls from 
the archives of the nome-capital Boubastos are today principally 
in the collections of Cologne and Vienna, with some 200 in 
Athens and some fifty at Duke University; found in the 1960s, 
they have recently begun to be published. The fragments from 
the archives of ancient Thmouis, capital in Roman times of the 
Mendesian nome, found between 1892 and 1906, are divided 
almost equally between Florence and Paris, with a few in Berlin, 
Geneva, and Manchester; some dozen of those fragments were 
published between 1903 and 1917.12 

au'tou~ d~ 'tTJV ioiav uv£px£O"8at. It is a pity that we cannot know more of the 
attendant circumstances of a letter writer's remark in P.Oxy. XIV 1668.17-21, 
<> TtYEf.lcOV UIlVT)uiav E1tEIl'l'EV Ev8uSE, Kat Ot)J(l;'tt q>6~0~ ouOE d~ EVEt. E<XV oilv 
8£A.£t~, tUEASe Ka'taq>povwv. 

10 WO 627, 631; cf JEA 23 (1937) 73. 
11 Some three dozen receipts for the payment of this tax are extant: to H. 

Braunert's list (jJP 9-10 [1955-56] 281) add SB VI 9604, nos. 18-20 (pace the 
ed. pro of Muller, see Excursus 2 below). 

12 These are listed in S. KAMBITSIS, ed., Le Papyrus Thmouis 1, colonnes 
68-160 (=Publications de la Sorbonne, Sir. "Papyrologie" III [Paris 1985]: 
hereafter 'Kambitsis') 2. P.Bub. I (=Pap.Colon. 15) was published in 1990. 



LEWIS, NAPHTALI, A Reversal of a Tax Policy in Roman Egypt , Greek, Roman and Byzantine 
Studies, 34:1 (1993:Spring) p.101 

NAPHT ALI LEWIS 105 

The new evidence on the merismos anakechorekoton is 
found in P. Thmouis 1, published with exhaustive commentary 
and awesome mastery of its myriad details by Sophie Kambitsis, 
who devoted twelve years of assiduous effort to the decipher­
ment, sequential arrangement, and elucidation of 109 fragments 13 

that are anything but easy to read. Her edition includes columns 
68-160 of the original roll; columns 10-67 and 161-77, as yet 
unpublished, are in Florence. 

(a) The Policy Change. The matter that concerns us first 
appears at col. 70, line 16-col. 71, line 10. In a village that 

had now fallen from 150 men to only forty-five men, of 
whom thirty-four had fled because impoverished and only 
eleven remained, the village clerk [komogrammateus] took 
off the books the amount of unpaid taxes levied on the 
fugitives; and he further stated that Annius Syriacus, the 
then prefect [of Egypt], when he conducted his conventus 
for the nome in year 3 [= 162/163], decided in similar cases 
that those who remained should pay [only] the amount 
levied on themselves.14 

That decision of the prefect is referred to twice again in 
P. Thmouis 1. At 124.18-125.2 we read of two villages that, 
"formerly populous, have now fallen to only fourteen men, of 
whom ten have fled and only four remain behind, and the 
amount of tax applicable to them must be collected in accor­
dance with the decision in year 3 by Syriacus of excellent mem­
ory." And at 152.13-153.4 we read of another komogrammateus 
who reported that with regard to certain taxes deferred [by his 
predecesor as prefect] "Annius Syriacus, in the conventus that 
he conducted for the nome in year 3, subscribed as follows: 
,[Those remaining] are to pay the amount of tax levied upon 
them, and the others are to pay the rest whenever they return 

13 Thirty-three of them in Florence, the rest in Paris (see Kambitsis). 
14 70.16-71.10: ano avopwv] pv ... vuv de; IlOVO'Ue; lCU[tTlV'tTj1CEVat] ~yo(pae;) liE, 

alP' bv 'tou<; [Ilev AO i:~uO]8Ev~OUV'tu<; aVUlCEXro(p11lCEVat) [IlOVO'U<; Or ill O'UV]­
EO'tuoBm, flCQ1'>[<ptOEV 'to btt~UA(AoV) 'tOtc; avalCElxro(p~lCaOt) lCat npooE811lCEV 
"Avvtov L'UptalCOV 'tOY T]YEIlOVE\'>[O]uv'tu f<P' olloirov lCElCptlCEVat 'tiP [y] (['tEt), 
O'tE 'tOY VOIl(OV) OtEAoyit;E'tO, 'tou~ O'U~IlEVOV't(U<;) 'to f7tt~UA(AoV) uu'tOt<; 'tEAEtV. 



LEWIS, NAPHTALI, A Reversal of a Tax Policy in Roman Egypt , Greek, Roman and Byzantine 
Studies, 34:1 (1993:Spring) p.101 

106 REVERSAL OF A TAX POLICY IN ROMAN EGYPT 

home'.» 15 In all these instances the references are to several 
taxes, none of which is a merismos. 

Though those three reports of Syriacus' decision are some­
what differently worded, their sense is the same, their import 
unmistakable: with that decision, made in response to a crisis of 
depopulation, the remaining taxpayers were required to pay 
only their own taxes; they ceased to be collectively liable for the 
taxes of the village's fugitives; the taxes of the latter were taken 
off the books of current receivables and classified as amounts to 
be collected from individual fugitives when (or if) they returned 
home. 

At 77.10-78.16 the effect of Syriacus' ruling is recorded in 
detail with respect to the village of Psenathre in the Ptenchat 
toparchy: 

The komogrammateus reported that in year 16 of the deified 
Hadrian [= 131/132] the number of men in the village 
registered in the census was 319, but now the number of 
poll-tax-paying men has fallen to only ten, of whom eight 
have fled because, he says, being few they can not pay the 
whole [village assessment of 432 dr. 3 ob. 3 ch.]. ... Of that 
sum the komogrammateus reported that there was due to be 
collected from the two men remaining 9 dr. 4 ob. (for, he 
says, at the time when it was first settled that the aforesaid 
sums be paid, there was levied on each man 4 dr. 5 ob.), and 
he took off the books the balance of 422 dr. 5 ob. 3 ch. of 
silverY' 

Without the explanatory information a similar calculation for 
another village appears at 126.6-127.2. 

15 124.18-125.2: 7taAat 7tOA:\laVOpou~ ouoa~ vuv d~ ~6vou~ t8_ Ka'tf)v'tf)l.£Vat, 
E~ Cilv avalCEXo>pf)K(EVat) iivo(pa~) t imoAi:t7tEoBat oc ~ovou~ 0, Cilv 'to a l.pouv 
OcpdA(OV) np<lx8(Tlvat) aKoW80>~ 't[ o'i~lK[pt8(E'i(n) 'too y (£'tEt) U1to] 'tou apio­
'tf)~ ~Vtl[~f)~ LUpt]aKou. The Greek text of i 52.13-153.4 'is at n.29 below. 

16 77.10-78.16: o· lC~'lOypa(fl.~a'tEu~) E0tlA{rooEV) £!Vat 'tqll~ (£'to) [8cou'AO]­
ptavou 'tou~ E7t' al)'tTl~ ava[ypacpo~E]vOU~ ~vopa~ itO, vuv OE [d~ ~6vo]~~ KU't­
f)V1:T)K{EVat) ofl.oA6you~ ... iivo{pa~) t, acp' wv ava[KEXo>pllK{EVat) ii]vopa~ li Oux 
'to, cpf)oi, ~it [ouvaoSat O]Atyou~ ov'ta~ 'to nav dO[cpEPElV] ... E~ Cilv 0 lCO>fl.O­
ypa{~~a'tEu~) EOtlA{rooEV) OcpdAOV a1tat'tctoSat napa 'toov OuvEo'tono>v av­
opoov ~ (op.) SF, 'tql yap xpOvcp, cpf)Ot, 0'tE 1tpOnco~ Eo'ta8f) otooo8at 'ta 1tpoKd­
~£va K£cpaAata btt~£~Af)KEVat hao't'!> avopt (op.) o(1t£v'tCOp.). Kat 'ta~ 
AOl1t(a~) £KOUcptOEV apy(upiou) (op.) UKP(1tEV'tCOP.)xT. On ofl.oMyou~ see n.25 
infra. 



LEWIS, NAPHTALI, A Reversal of a Tax Policy in Roman Egypt , Greek, Roman and Byzantine 
Studies, 34:1 (1993:Spring) p.101 

NAPHTALI LEWIS 107 

We can, as it happens, date Syriacus' decision to a precise 
month. M. Annius Syriacus was prefect of Egypt in 161-164,17 
The prefect's conventus for the nomes of the eastern Delta was 
held annually in January.18 In year 3 (of Marcus Aurelius and 
Lucius Verus) that would correspond to January 163. That is 
when, as we learn from P. Thmouis 1, the Roman government 
gave up demanding the taxes of the fugitives from those who 
remained behind, and it is thus no accident that the last extant 
receipt for a payment of the merismos anakechorekoton dates 
from 161. 

At 120.13-21 P. Thmouis 1 cites some komogrammateis of 
166/167 as reporting that certain villages were completely de­
populated, and that "at his conventus Syriacus of excellent 
memory had canceled such taxes." 19 Although that order (as the 
editor carefully notes [45]) was issued "pendant Ie conventus 
d'une annee non indiquee,» it fits hand-in-glove with, and was 
very likely part of, Syriacus' above-discussed ruling of January 
163: where a few taxpayers remained, Syriacus ended their 
collective liability for the taxes of the fugitives; where no tax­
payers remained-i.e., where the depopulation was total-the 
tax cancellation was total. In all likelihood, then, this last was also 
part of the decision of January 163. Alternatively, it could have 
been a supplement, or 'follow-up', issued at the conventus of 
January 164. After that Syriacus was out of office. 

As it appears in P. Thmouis 1, Syriacus' policy change was not 
promulgated in a general edict addressed to all of Egypt, but was 
issued in a conventus in response to specific crisis situations in 
the Mendesian nome of the Nile delta region. But, as noted 
above, some komogrammateis of that region referred to Syri­
acus' having made the same ruling in "similar situations,» and 
some of those situations may have been in other nomes, e.g. in 

17 Cf G. Bastianini, "Lista dei prefetti d'Egitto dal 30' al 200P," ZPE 17 
(1975) 295f, and "Lista dei prefetti d'Egitto cial 30' al 200P. Aggiunte e cor­
rezioni," 38 (1980) 82. 

18 The annual schedule for the prefect's conventus in Egypt was detailed by 
U. Wilcken, ArchPF 4 (1908) 366--433, 6 (1920) 373-76, further refined by G. 
Foti Talamanca, Ricerche sui processo nell' Egitto greco-romano I (Milan 
1974). 

19 120.13-21: <j)l)OUVtEe; t[ae;jKwllUe; mlAat 1toA'Uu[v8jpo'Ue; ououe; vuv de; 0/"'1-
[youe; 8j/"'we; 1tEptEOtUo{)udoue; KUt Ejv aVUXWP~OE;' dVeXt, [1tPOOC8TlKjUV 8£ KUt 
r'UptaKov [tOV aplot]Tle; ~V~~TlC; E1tt tOU [OtuA.oytO'~]ou ta 8~otu a1tOA.e/..,U[KCvUt. 



LEWIS, NAPHTALI, A Reversal of a Tax Policy in Roman Egypt , Greek, Roman and Byzantine 
Studies, 34:1 (1993:Spring) p.101 

108 REVERSAL OF A TAX POLICY IN ROMAN EGYPT 

Upper Egypt, whence the extant receipts for the tax run, as 
noted above, only to 161. In any case, there is evidence of the 
same or similar dispositions in another nome not many years 
later. 

On the day corresponding to 21 February 168 the prefect Q. 
Baienus Blassianus wrote to a strategos in the Arsinoite nome: 

The taxes of the destitute fugitives customarily collected 
through a merismos I grant for the present to be suspended 
in order that those in flight may return to their homes and 
those still there may be able to remain .... 20 And you must 
... carefully ascertain who in reality are fugitives .... For, even 
if a postponement of the tax collection has become necessary 
for the present, care must be taken so that hereafter too the 
indigenes are not burdened with the taxes of non-fugitives.21 

Like Syriacus' ruling, this suspension of the merismos sounds 
like an ad hoc response to a particular circumstance: note espe­
cially avuYl(uiu £YEVE'tO, a stock phrase in imperial and prefec­
torial edicts. 22 Note also the repetition of 1tpo<; 'to 1tUPOV, empha­
sizing the prefect's expectation that the merismos would be re­
imposed after a brief respite (1.l1tEpSEOl<;, a term that, perhaps 
significantly, does not occur in P. Thmouis 1). Whether it was 
later reimposed will be discussed in Excursus 2. 

(b) The Sequel. In the Mendesian nome, as reflected in 
P. Thmouis 1, Syriacus' policy was continued by his successors 
in the prefecture for about ten years. In one village, 

pursuant to his report to the then prefect Blassianus that 
the village had been attacked and burned already in year 7 
[=166/167], the komogrammateus suspended the taxes due, 
adding that the male population in the village had fallen 
from a large number to only two, and they had fled. Conse-

2D That is, they will not have the additional burden of paying the fugitives' 
taxes. As appears from this and the previously quoted passages, 'suspend' in 
this context is tantamount to 'cancel': cf ZPE 38 (1980) 251£. 

21 SB XIV 11374: 'tal £7tl1c£cpaAna 'twv uval(£xroptl'Co'trov U1tOProv O'UVtlSro<; 
[EI( ~]£plO'~OU dO'cpqx>~£va O'tlVxropwt 1tpo<; 'to 1tapov [i:1t]tol(£Siivat (1. -O'X£-), 
tva or EV uvaxroPtlot QV't£<; E1taVfA9roat d<; 't~v oil(tav I(at Ot QV'tE<; Otl~~fV£tv 
SUVOV'tat (I. -roV'tat) ... 1tPOOtll(£t Sf I(at Of ... E1tt~£AW<; ES£'tasd tv] 'ttV£<; dOt 
Ot 'tip OV'tt uval(£xropl1l(o't£<; ... d yap 1tpo<; 'to 1tapov (lvuyicaia' £ytv£'to \)1tfP­
SEat<; 'tTl<; dO'1tpa'~EW<;, aMo\) 1tpovOll'taiov ( 1. -tOY) £o'ttV ~l1Sf uo't£pov 'tou<; E1tt­
XroptO\)<; l(a~[a~a]p£toSat dO'cpopat<; 'tWV OUI( uval(£xropl1l(O'trov. 

22 E.g. SB I 3924.10 (edict of Germanicus, 19), Hibis 4 (GGIS II 669, IGR I 
1263.8, 19, 41 [edict of Ti. Julius Alexander, 68]). 



LEWIS, NAPHTALI, A Reversal of a Tax Policy in Roman Egypt , Greek, Roman and Byzantine 
Studies, 34:1 (1993:Spring) p.101 

NAPHTALI LEWIS 

quently, deferral continued from then [year 8] to year 10 
[=169/170l 

109 

The nome secretary adds, in excellent bureaucratese, «Accor­
dingly, [the taxes] have been suspended [by me] for year 11, 
since ... no [prefectorial] decision relating to them has been 
made known to me."23 Similar language follows (116.2-16, 
117.3-17) with respect to two other villages. Further, several 
komogrammateis report that 

they have taken off the books the taxes of the men regis­
tered in their villages, because in former times ... the villages 
had been populous but now some were completely deserted 
while others had fallen from a large number of men to a 
few, the majority of whom had fled because impoverished; 
and they further stated that Bassaeus Rufus, vir clarissimus, 
in year 9, when he conducted his conventus for the nome, 
canceled similar taxes for other villages of the nome.24 

That was the conventus of January 169. Those suspensions also 
continued into year 11 (=170/171). 

A modification of the policy came at the conventus of year 13, 
that is in January 173. On that occasion the prefect C. Calvisius 
Statianus ruled, with reference to some twenty villages, that a 
portion of the taxes taken off the books in prior years should be 
restored to the tax rolls as collectible. 25 In one instance, for 
example, the reinstatement amounted to 128 dr. 2 ob. 4 ch. from 
a total of 922 dr. 1 ob. 3 ch., in another 73 dr. 2 ob. from 300 dr. 4 

2J 114.6-20:3) K(OJ.loypa(J.lJ.lu'td)~) ... cp"aa~ 'tTtv KroJ.l(llV) cepoSou CJ.llt£ltPlla9at 
£n altO 'OU t; (£'t"O"u~) aKOAou9co~ 'Ot~ BAaaatavip .ip lryEJ.lovEuaav.t ypu­
(<pEtcrl), OlCcr'tElAEV [the taxes ..9f year 8] ltpocr9El~ 'tou~ ev athn avopa~ a1tO 
1tAEiovo~ apl8(IlOU) Ei~ J.lOVOU~ ~ Ka'l1vulK£Vat, oucr1tEP avanxcopllKCVat. [V8EV 
Ka.' aKoAou8(iuv) au'tOu £K'tO'E J.lCXpt 'tou t (£'tOu~) CltEaX£[911]' ltap' 0 Kat 
ltpO~.o ta (£'to~), eltElKa[8on]ltpOKEt't"at ouSi:v w~ 1tEp[l au.wv] Kpt9i:v epaVE­
pov J.l0l cy£v[ no], Staa~tAAE't"at. 

2A 76.10-77.2: ncpUAala ... ev 'Ot~ C1tt .wv K(oJ.lWV avaYp'aepoJ.lcvot~ avopum 
eKou<ptcrav e1tt .ip 'Ot~ 1tUAat xp6VOt~ .... a~ KroJ.la~ 1tOAUUVOPO\)~ dVat, vuv 0[1:] 
.a~ J.l£v 't£AEOV eYAEAolltCVat .a~ 0[£] alto 1tAdovo~ apt8J.lou avopwv Ei~ OAi­
youe; Ka'l1v'l1KEvat, aep' dlv .ove; ltAEia'OUe; e~ua9Ev"oav[ .u~] gvanxcoPll­
KEVat, ltpooE9T)Ka[ v 01:] Baooa'lov • Pouepov .ov AaJ.lltpo.[ mov] .ip 9 (E".Et), on 
.ov VOJ.lOV oU:Aoyi[ t;E'O]' 'ta 0J.l0l[ a t1tt] hcpcov KCOJ.lWV 'tou VOJ.l[ ou 1 altOAEAUKE­
vat. 

25 E.g. 73.19-22: .a~ Ei~ OJ.lOA(OYOV) altoKa.ao.(a9Eiaae;) ... w~ Kpte(Eiaa~) 
U1tO Lwnavou tv ~ e1to111cra.o .ou VOJ.lou OlUAoYlcr(J.lip) 'tip tY (£'tEt) [a]p­
r(upiou) (opaXJ.lae;) (numeral). For other instances see the Index General 5.'1). 

OJ.lOAoyOV (p.187); the related sense of OJ.lOAoyoe; as poll-tax payer occurs at 77.13 
(quoted above n.16); cf ChrEg 49 (1974) 162. 



LEWIS, NAPHTALI, A Reversal of a Tax Policy in Roman Egypt , Greek, Roman and Byzantine 
Studies, 34:1 (1993:Spring) p.101 

110 REVERSAL OF A TAX POLICY IN ROMAN EGYPT 

ob. 3 ch., in a third 780 dr. 4 ob. 6 ch. from 1883 dr. 3 ob. 1 ch. 26 

In no instance does the amount reinstated exceed forty per cent 
of the total suspended; in one instance it is as little as six per cent. 

In none of those instances is a reason stated for the prefect's 
decision. Kambitsis suggests (30) two possibilities: (1) the money 
came from the confiscation and sale of property abandoned by 
the fugitives; or (2) it came from fugitives who returned home, 
able to pay their arrears. Of these, (1) is the less likely: it is dif­
ficult to find that imputation in the Greek of the references to 
Statianus' ruling, and we know that many (if not most) of the 
fugitives left no taxable property behind.27 Suggestion (2), while 
conjectural, is a priori the stronger possibility; if confirmatory 
evidence ever appears, it would indeed imply that by the be­
ginning of the year 173 "la crise qui sevissait dans Ie nome etait 
en train de s'attenuer.» 

However that may be, we can be clear about what Statianus' 
action was and was not. It was a reimposition-for a reason or 
reasons at which we can only guess-of a small fraction of the 
total of taxes defaulted by fugitives, taxes whose collection had 
been suspended by the prefects of the preceding ten years in 
continuation of the policy instituted by M. Annius Syriacus. In 
the several references to Statianus' action there is not the 
slightest implication that the reimposed taxes were to be 
collected by reactivating the merismos anakechorekoton. In 
other words, Statianus' action of January 173 was not a return to 
the principle of collective responsibility. In fact, the language 
points in exactly the opposite direction. In each instance 
Statianus ordered the reinstatement of a precise sum, itemized 
to the last chalkous. Those can only have been the amounts 
owed by designated individuals, presumably returning fugitives. 
The purpose of the merismos anakechorekoton was to make 
good the totality of the tax shortfall caused by the disappearance 
of the fugitives, not the small fractions of Statianus' order. In 
short, Statianus did not reinstate the merismos anakechore­
koton. 

26 73.18-22,80.14--18,103.3-104.4. 
27 This is freqently emphasized; cf e.g. the document quoted above 2f; also 

P.Oxy. II 251, 252 (W.ehr. 215),253; XXXIII 2669; P.Mich. X 580. 



LEWIS, NAPHTALI, A Reversal of a Tax Policy in Roman Egypt , Greek, Roman and Byzantine 
Studies, 34:1 (1993:Spring) p.101 

NAPHT ALI LEWIS 111 

Excursus 1: Causes of the Depopulation 

While avaxwPTlcrt<; was a chronic phenomenon in Roman 
Egypt, there were periods when extreme distress led to extreme 
depopulation of the rural villages. 28 One such period has long 
been known to have occurred in Marcus Aurelius' early years. 
With P. Thmouis 1 we acquire a much clearer picture of those 
events and of their dates. 

(A) Crop Failure. Syriacus' policy change of January 163 came 
after a few years of tax deferrals granted by his predecessors. 
This is stated quite unambiguously in P. Thmouis 1 at 152.12-
153.18: 

The komogrammateus reported in year 23 [of Antoninus 
Pius=159/160] that [certain taxes] had to be deferred until 
prefectorial cognizance owing to the abandonment of one 
village and because in the others, formerly populous, only 
two men are registered in the census and they have fled. As 
for the deferred sums, in the conventus that he conducted 
for the nome in year 3 Annius Syriacus subscribed as fol­
lows: "[Those remaining] are to pay the amount of tax 
levied upon them, and the others are to pay the rest when­
ever they return home. "29 

The text goes on to say that the suspension of the fugitives' taxes 
continued to year 11 (=170/171). 

It thus becomes clear that severe depopulation began in the 
Nile delta even before the accession of Marcus Aurelius. The 
cause of the depopulation in those years appears at several other 
places in P. Thmouis 1. At 109.1-9 we read of taxes suspended 
on a vineyard that "the komogrammateus reported had for the 
first time in year 23 of the deified Antoninus Pius become 
chersos [dry]," i.e., land unproductive because beyond the 

28 Although severe, total depopulations, such as those evidenced by P. 
Thmouis 1, were the result of occasional crises, avaxwPllcru; was a chronic and 
endemic phenomenon that simply peaked in those crises. As characterized by 
Kambitsis, avaxwPllcru; was "un fait frequent et banal" (28). See Excursus 3 
below. 

29 152.12-153.4: 0 lCwfloypa(flflatE1)~) tip iCy (£tEt) £SljAwcrev O<pctAEtV 
8tacrtaAllvat axpt i)yeflOvtxll~ 8tayvwcrew~ 8ux to ti}.v fl£V lJPllfliOO8at, btct 81: 
'tow £tEPWV 1taNn 1toAuav8pw1twv oucrillv fl6vou~ ~ avaypa<pccr8at, OU~ lCat 
avm':£XW(Pllldvat). tillv 81: lCe<paAalWV Stacr1:aA(EvtwV) "Avvw~ LUPtaVlC6~, 
£v cp £1tOtljcrato tip y (£1:Et) tOU VOflOU 8taAoytcr(flip), {J1tEypmjlev OUtW~' ... to 
[£]1tt~aAAoV aUtot~ 1:eAe<?"~['t]wcrav, :[0] 81: Aot1tOV oi UAAot otav £1tavEA8wcrtv. 



LEWIS, NAPHTALI, A Reversal of a Tax Policy in Roman Egypt , Greek, Roman and Byzantine 
Studies, 34:1 (1993:Spring) p.101 

112 REVERSAL OF A TAX POLICY IN ROMAN EGYPT 

reach of the fertilizing Nile flood. 30 That tax suspension also con­
tinued to year 11. Another such vineyard case is recorded at 
110.1-11. 

At 156.17-157.7 there is a statement regarding certain parcels 
of land normally devoted to growing cereal crops: because they 
were found on inspection in year 22 (=158/159) to be chersos 
the taxes on them had been likewise deferred, in most cases 
continuously also to year 11. 

Lastly, at 108.16-20 we read: "The taxes on a vineyard ... were 
suspended in year 1 [160/161] of our lord Aurelius Antoninus 
Caesar and the deified Verus upon its becoming chersos. "31 It 
thus seems clear that the basic-though probably not the on­
ly-cause of the depopulation in the years leading up to Syri­
acus' decision reversing the practice of collective tax respon­
sibility was a series of inadequate Nile rises and consequent crop 
failure or insufficiency. 

(B) Plague. Early in 166 the ravages of a plague forced Lucius 
Verus to abandon a victorious campaign in Parthia and return his 
army to its base in Syria. Perhaps in extrapolation of Ammianus 
Marcellinus' report (23.6.24) that the plague swept the Roman 
Empire spreading its contagion everywhere ad usque Rhenum 
et Gallias, most historians of the nineteenth and twentieth 
centuries have tended to maximize-often in purple prose-the 
plague's deleterious effects upon the Empire, some even de­
picting it as causing death and devastation so extensive that "the 
ancient world never recovered from the blow inflicted upon it 
by the plague which visited it in the reign of M. Aurelius." 32 Not 
many, I think, still regard that plague as such an historical water­
shed, but it was surely an epidemic that spread to many parts of 
the Empire, including Rome itself. 

Did the plague penetrate Egypt? Given the proximity to Syria, 
one would expect that it did. But until now the only ancient 
source to make mention of Egypt in that connection has been an 

30 109.1-9: 1tUPE'tESll 'tOY KW!l0YPU(!l!lU'tEU) 1tpO<Ht£<PWVllKEvat 1t po)'tws 'tiP 
KY (e't£t) SEQU AlAtou 'AV'twvtvou 'tTjv o.!l1t£A(OV) ... KEXEpa&aSat ... (K'tO'tE !lEV 
o-ov HillAwell -:u 't£AEO!lU'tU E1tEOxlloSat. . . 

31 108.17-20: al.l.1tEAOU ... 'tu 'tEAEO!lU'ta 'tiP a «('t£t) AUPllAtOU 'Av'twvtvOU 
Kuioupos 'tOU Kupiou KUt SE,?U OUr,pou E1tl:OXESll E1tt 'tiP K~X£PO&>oeut. 

32 The most thorough review of the ancient sources is still that of J. F. 
Gilliam, "The Plague under Marcus Aurelius," AJP 82 (1961) 225-51; but his 
evaluation of the papyrological evidence must now be revised in the light of P. 
Thmouis 1. The quotation is at 225, citing B. G. Niebuhr. 



LEWIS, NAPHTALI, A Reversal of a Tax Policy in Roman Egypt , Greek, Roman and Byzantine 
Studies, 34:1 (1993:Spring) p.101 

NAPHTALI LEWIS 113 

obscure first- or second-century historian, Crepereius Calpur­
nianus, and he, following his Thucydidean model, describes the 
plague as originating in Ethiopia and spreading thence through 
Egypt into Parthia, where it remained-good riddance. 33 Nearly 
a century ago Wilcken acutely suggested that the severe depopu­
lation recorded in the early-published carbonized fragments 
(supra n.12) resulted from the plague. A quarter-century later, 
with increasing evidence of UVUxwPTl(Ju; in published papyri and 
none of epidemic, Wilchen retracted his suggestion; it con­
tinued, however, to be espoused by others.34 

A passage in P. Thmouis 1 now tells us expressis verbis that the 
plague did indeed spread into Egypt-at least into the Delta 
region-and decimate the population. From 104.9-18 we learn 
that in year 9 (=168/169) a komogrammateus "took off the 
books [certain] taxes, declaring that most of the inhabi tan ts of 
the village had been killed by the rebellious Nikochites when 
they attacked and burned the village, and others had died of the 
pestilential condition, and the rest, very few in number, had 
fled. "35 

(C) Rebellion. A brief notice in Cassius Dio (72.4.1f) is our 
principal literary source for the revolt of the Bucoli, who 
launched their attacks-even against Alexandria, if we are to 
credit Dio-from their hideouts in the swamps of the Nile delta. 
Dio appears to place the outbreak of the uprising in 172/173. 
P. Thmouis 1 now reveals that the attacks on the villages of the 
Delta began as early as 166/167. 

3J FGrHist 208 with commentary. The statement is relevant to our concern 
irrespective of the controversy over the historicity of Creperius Calpurnianus et 
al., on which see e.g. C. P. Jones, Culture and Society in Lucian (Cambridge 
[Mass.] 1986). I thank Everett Wheeler for this reference. 

H Notably H. Braunert, Die Binnenwanderung (Bonn 1964) 166; also A. E. 
R. Boak, "The Population of Roman and Byzantine Karanis," Historia 4 
(1955) 250. Wilcken's retraction of his 1903 conjecture came in ArchPF 8 (1927) 
311. For further details and bibliography see P. Thoumis 1,29 n.3. Relevant 
and possibly relevant documents from Egypt published since Gilliam's article 
(supra n.32) have elicited a fair amount of comment, notably from G. 
Casanova; see, most recently, "Altre testimonianze sulla peste in Egitto. 
Certezze ed ipotesi," Aegyptus 68 (1988) 93-97 with bibliography. It should be 
noted that Casanova's hypotheses and reconstructions are not universally 
accepted. 

35 104.9-18: 0 uinae; 1(WJ.10YPU(J.1J.1UtEUe;) ... E1(OUcptOTV cpi)oue; tOUe; 1tAdotoue; 
trov a1ta tile; lCcOJ.1TJe; avnpilo9m U1ta trov avooiwv NnlCwlC£nrov E1tEA96vtwv 
tU lCWJ.1U 1(ul EJ.11tpTJoavtwv uini)v. lCul aAAO'I)C; tip AOtJ.1tlCip 1(utaO'tl)J.1un 
t£'tfA(EutTJ1(Evm). 1(U 1. tOUe; AomoUe; OAtyiotouC; onue; a1to(1tE)cpfuyEvm. 



LEWIS, NAPHTALI, A Reversal of a Tax Policy in Roman Egypt , Greek, Roman and Byzantine 
Studies, 34:1 (1993:Spring) p.101 

114 REVERSAL OF A TAX POLICY IN ROMAN EGYPT 

P. Thmouis 1 has three references to such attacks. (1) The first, 
referring to 168/169, is in col. 104, quoted just above in (B) of 
this excursus. The identification of those "rebellious Niko­
chites" with the Bucoli was made by M. Manfredi citing Achilles 
Tati us 4.12.7f, where Nikochis is called the headquarters of the 
Bucoli. 36 (2) At 114.6-10 we read of a statement by a komogram­
mateus "that the village was attacked and burned already in year 
7 [=166/167], as was reported to Blassianus the then prcfect."37 
While the Bucoli are here not cited by name, the parallelism 
with the language of col. 104 and the temporal proximity leave 
little doubt that they were in fact the attackers. (3) At 116.2-11 
we read "that of the fisherman some were killed by the rebel­
lious Nikochites who attacked the village (as reported ... to 
Blassianus the then prefect) and [the number of the rest] fell to 
only five men, [so that the komogrammateus] took the share of 
taxes of the missin g twenty-six off the books." 38 The mention of 
year 8 that follows, combined with the reference to a previously­
rendered report to Blassianus (who was prefect in 167-168), 
would seem to indicate that this village, like the one of (2) above, 
was attacked in 166/167. 39 

Excursus 2: Is the M erismos Anakechorekoton 
in Evidence after Marcus Aurelius? 

The question is really bipartite: (1) After the the suspension 
was the tax later reimposed? (2) Was the tax collected without in­
terruption in some nomes e.g. in Upper Egypt (far from the 
raids of the Bucoli and, as far as we know, from the plague)? 

36 Cf. P. Thmouis 1.104.13n. (p.99). 
37 114.6-1Q: 0 KWlloypa(lllla'tdJe;) ... q111aae; 'tf]V KWIl(T]V) E<jJOOO'U EIl1t£1tPl1oElat 

E'tt a1to 'tau I; (£-ro'Ue;) (lK:OAoU8we; 'tOte; BAaacrtavip 'tip i]Y£llov£uaav'tt ypa(rtcrt). 
38 116.2-11: 0 KWlloy[pa(lllla'tdJe;)], <jJT]aae; 'toov all<jJt~oA(£WV) 'tOue; 1l£V 

't£['t£jA(£'U'tT]K£Vat) 'tOue; of. aV11Pl1a8at U1tO 'toov avoa£tWv NtKWXEt'tooV 
E1t£A8ov't[wjv ... Kat 1t£pt 't11e; E<jJOOO'U [ypja(<jJl1vut) ... 'tip i]Y£1l0vEUaaV'tl 
BAaaatavip Kat de; 1l0vo'Ue; Ka'ta~'tl1a(at) de; £, 'too. Ota<jJ£pov'twv K~ 'to Il£p<;e; 
EKoU<jJt<JEV . 

39 So too Kambitsis 115 ad 115.21-116.18. In the following year, as we learn 
from 99.3-11, the remnants of the population of three villages, apparently 
because they were suspected of having sided with or helped the rebels, "were 
killed ... by the military force dispatched, and the villages are now [170/171] 
totally deserted" (uv11Pl1a8at 'tip 11 (c'tEt) U1tO 'tl1S 1t£1l<jJ8d'!T]s a'tpa'tlw'ttx:l1s 
OWUIl£We; Kat OA£pT]Il0'Ue; ctVat 'tae; KWllae;). 



LEWIS, NAPHTALI, A Reversal of a Tax Policy in Roman Egypt , Greek, Roman and Byzantine 
Studies, 34:1 (1993:Spring) p.101 

NAPHT ALI LEWIS 115 

The evidence on merisomoi in general is far from clear-cut. 
There are, for example, dozens of receipts for taxes called 
merismoi without any following qualifiers. S. L. Wallace long 
ago contemplated and rejected as unlikely the notion that those 
(Sc. or some of them) might record payments for the merismos 
anakechorekoton. 40 That leaves for our consideration a single 
document, an ostracon from Syene in the Ashmolean Museum 
published as O.Tait 166 no. 21. 

Tait, following the lead provided by WO in half a dozen ex­
amples, read the payment as made un(Ep) /lEPlO"/l(OU) av8(puxv­
'toe;;) av(a)K(EXPUO"W/lEVOU). Those resolutions of the abbrevia­
tions were called into question by Wallace, who argued for re­
vising to av<>(pwv) av(a)K(ExwPTlKO'tWv).41 This was rejected by 
W. Miiller,42 but his arguments are far from compelling; the 
reading av<>puxv't(oe;;) of WO 151, which he cited as "ent­
scheidend,» was revised to avaKEx(wPTlKO'tWV) in the re-edition 
of the ostracon as O.Leid. 181. Recently, ignoring Miiller's inter­
vention, the late John C. Shelton demonstrated convincingly 
that Wallace had been right all along about the av8( ) avaK( ) 
receipts, and that Wilcken's century-old notion of a head tax to 
pay for regilding imperial statues was an insubstantial pipe­
dream, a will-o'-the-wisp pure and simple. 43 

Having discarded the editor's resolution of the key abbrevi­
ations, do we then have instead in the Ashmolean ostracon a 
receipt for merismos anakechorekoton? If so, that would be 
decisive for the question of this excursus, because that receipt 
was very likely (from the probable identity of one of the tax 
collectors) written in 194. The sticking point is the assumption 
that aVK was miswritten for avaK, and that not only once but 
twice. 44 Unicum ergo suspectum. It might conceivably be 
argued that writing aVK for avaK was a (phonetically induced?) 

-40 Taxation in Egypt (Princeton 1938) 167f. His argument from the tax rates 
or amounts, however, strikes me as less cogent than a simple a priori 
deduction. 

41 Wallace (supra n.40) 160f, followed by Braunert (supra n.ll). 
42 MODG 6 (1958) 4. 

43 "Ostraca from Elephantine in the Fitzwilliam Museum," ZPE 80 (1990) 
237f. Note especially, "The word avaxpuoow is not attested outside of 
abbreviations in this context. That is, Wilcken invented it for the sake of this 
tax." 

H A photograph of the ostracon obtained from the Ashmolean supports 
Tait's reading of avli av K in lines 6f. 



LEWIS, NAPHTALI, A Reversal of a Tax Policy in Roman Egypt , Greek, Roman and Byzantine 
Studies, 34:1 (1993:Spring) p.101 

116 REVERSAL OF A TAX POLICY IN ROMAN EGYPT 

vagary of this particular writer. But that is surely a shaky 
foundation of which to build a case, the more so as the rest of 
the text of the ostracon shows only a single instance of an 
omitted vowel, the second iota of the name Bienchis. 

In sum, the suspension of the collection-in other words, the 
cancellation-of the merismos ankechorekoton in at least parts 
of Egypt under Marcus Aurelius is amply attested. Its survival 
or revival under Septimius Severus is, on present uncertain 
evidence, at most a possibility. 

Excursus 3: Broader Implications 

If a man were called to fix the period in the history of the 
world, in which the condition of the human race was most 
happy and prosperous, he would, without hesitation, name 
that which elapsed from the death of Domitian to the acces­
sion of Commodus .... The forms of the civil administration 
were carefully preserved by Nerva, Trajan, Hadrian, and the 
Antonines, who delighted in the image of liberty and were 
pleased with considering themselves as the accountable min­
isters of the laws .... The labours of these monarchs were 
overpaid by the immense reward that inseparably waited on 
their success; by the honest pride of virtue, and by the 
exquisite delight of beholding the general happiness of 
which they were the authors (Edward Gibbon, The History 
of the Decline and Fall of the Roman Empire (1776) ch. 3. 

From Gibbon to the present that has remained the accepted 
view of the Roman Empire: a world of great, probably unprece­
dented, well-being for an unprecedented number of people 
during the first two centuries of the Pax Romana, with the 
decline (leading to the fall) commencing in the tumul tuous and 
distressed conditions of the third century. 

Brilliant synthesizer that he was of awesome quantities of 
varied and widely diffused primary and secondary sources-and 
that in the pre-computer age-Rostovtzeff worked comfortably 
within that conventional framework. Thus, in his treatment of 
avaxwPll(Jl<; (Roman Empire [1926] 528 n.50) we read: "The 
avaxwPll(Jl<; remained the characteristic feature of Egyptian life 
even in the comparatively happy period of the first and the 
beginning of the second century. It seems, however, as if the 
mentions of avaxwpl)(J£u; in this period may all be explained by 
exceptional circumstances." In the second edition (1957) Peter 
Fraser repeats this (677 n.52), citing PSI 1043 as possible addi-



LEWIS, NAPHTALI, A Reversal of a Tax Policy in Roman Egypt , Greek, Roman and Byzantine 
Studies, 34:1 (1993:Spring) p.101 

NAPHTALI LEWIS 117 

tional support. But the impact of the increasing documentation 
is not lost on Fraser, who adds, "Yet the ultimate cause was 
undoubtedly the policy followed by the Roman government in 
the exploitation of Egypt" (with a reference to Rostovtzeff's 
own article in JEconBusHist 1 [1929]).45 

Under Septimius Severus, as Rostovtzeff read the evidence, 
"the conditions of the land were very bad and ... uvuxc.oPTtO"£ts 
became a real plague" (599 n.15 1=712 n.15 2). By the middle of the 
third century, "under the pressure of these [anarchic] condi-
tions ... people fled from their places of residence ... [ to] a life 
of adventure and robbery in woods and swamps ... It was an 
everyday occurrence for a man to have his property sold up, to 
become a beggar, to flee from his place of residence" (424, 
428 1=476,480 2 ). 

In reality, from accumulated evidence, old and new, it has 
become clear that long before the third century UVUxwPllO"ts 
was not the occasional, crisis-generated phenomenon envisaged 
by Rostovtzeff (crises, of course, made matters worse), but a 
chronic element of life in Roman Egypt, "un fait frequent et 
banal" (Kambitsis 28). The unambiguous evidence of the 
merismos anakechorekoton is available for most of the second 
century from its earliest years, and the normal genesis of 
legislation makes it reasonable to suppose that the phenomenon 
of uvuxwPllO"ts had been around for some time (and growing) 
before the tax was introduced. 

"Coming events cast their shadows before.» In one matter 
after another ancient historians in our own time have discerned 
the seeds of third-century decay in the second-century pros­
perity. The history of the merismos anakechorekoton adds 
another item of supporting evidence for that view, at least for 

<15 In Klio 75 (1993) 306-20 (a reference that lowe to Everett Wheeler), S. 
Link reverts-apparently unwittingly-to Rostovtzeff's position. Anachoresis , 
he posits, was in the first century not chronic but only sporadic, and often a 
canny scheme of tax avoidance rather than a sign of economic distress. The 
evidence? "Wahrend des ganzen ersten Jahrhunderts finden sich nur wenige 
Hinweise darauf" (307). It ought not to be necessary to point out that such 
'deduction' ignores the probability that the "wenige Hinweise" result from the 
chance of the finds, which, for almost any subject of investigation, has 
produced a vastly greater documentation for the second century than for the 
first. 



LEWIS, NAPHTALI, A Reversal of a Tax Policy in Roman Egypt , Greek, Roman and Byzantine 
Studies, 34:1 (1993:Spring) p.101 

118 REVERSAL OF A TAX POLICY IN ROMAN EGYPT 

Egypt.46 This is hardly the place to enter into a discussion of that 
general view; but as a symbol of the general development we 
may perhaps cite the fate of Marcus Aurelius, the 'philosopher­
king' who aspired to a life of quiet meditation as the summum 
bonum but was compelled to spend most of his Principate in 
military activities protecting the Empire's frontiers. 

CITY UNIVERSITY OF NEW YORK 

August, 1993 

46 In the first half of this century most treatments of Roman Egypt em­
phasized what was conceived to be its uniqueness among the provinces of the 
Empire. That view, engendered by the sudden emergence of thousands of 
papyri with uniquely detailed information and bolstered by a tendentious 
reading of Tac. Rist. 1.11.1 and Ann. 2.59.4, has given way in recent decades to 

the realization that much of the 'uniqueness' is like! y to be specious: it may 
look that way because we do not have comparably intimate information from 
other provinces. As Alan Bowman put it (jRS 66 [1976] 161), 

If Egypt is in some respects atypical we must not only remem ber that 
other provinces also had peculiar features (which might induce us to regard 
them as atypical, if we knew as much about them), but also ask ourselves 
what we might reasonably expect to be able to say about 'typicality' in the 
Empire. The important thing is to treat the evidence on its merits and to 
realize that, whilst the papyri may reveal details which are not literally 
applicable to provinces other than Egypt, they may, sanely applied, 
illuminate administrative, social and economic features of the Empire as a 
whole.