214 The Imperial Government
his native city, honouring Claudius’ doctor, C. Stertinius Xenophon,who
is also described as holding the post ofepi tōn Ellēnikōn apokrimatōn(to do
withpronouncements,orreplies,inGreek).Howexactlyweshouldtrans-
lateapokrimais not clear, but hardly matters in this context, as the sphere
ofactivityisclearenough.ItisequallysointheentryinaByzantinelexi-
conwhichdescribesanotherGreek,ofthesecondhalfofthefirstcentury,
Dionysius of Alexandria, as ‘‘in charge of theepistolaiandpresbeiai[embas-
sies] andapokrimata.’’FromthereignofHadrianonwards,theroleiswell
attested,andinparticularitisknowntohavebeenheldbyanumberofthe
fashionableandfamousGreekoratorscelebratedintheLives of the Sophists
ofPhilostratus,writteninthefirsthalfofthethirdcentury.Thatfactalone
wouldtendtosuggestthattheirrolerelatedtotheformulationofimperial
lettersincorrectGreekstyle.And,infact,thisfunctionismadequiteex-
plicitinwhatPhilostratussaysofoneofthesemen,Aspasius,whoheldoffice
intheearlythirdcentury.TheotherPhilostratusmentionedwasthewriter’s
nephew(Vit. Soph.2,33):
TheepistlecomposedbyPhilostratuscalledHowtoWriteLettersisaimed
atAspasius,whoonbeingappointedimperialsecretarywrotecertain
lettersinastylemorecontroversialthanissuitable;andothershewrote
inobscurelanguage,thoughneitherofthesequalitiesisbecomingto
anemperor.Foranemperorwhenhewritesaletteroughtnottouse
rhetoricalsyllogismsortrainsofreasoning,butoughttoexpressonly
hisownwill;noragainshouldhebeobscure,sinceheisthevoiceof
thelaw,andlucidityistheinterpreterofthelaw.Inthecourseofthesecondcentury,thedivisionoffunctionsnaturallyled
totheothersecretaryconcernedwithlettersbeingdescribedasab epistulis
Latinis.Itwouldnotbeunreasonabletosupposethattheirrespectivefunc-
tionswerealreadynotverydifferentfromthoselaterdescribedintheNotitia
Dignitatum(ListofOffices)ofarounda.d.400:
Magister epistolarum:dealswithembassiesfromcities,consultationes[en-
quiriesbyletterfromofficials]andpreces[petitions].Magisterepistolarumgraecarum:thoseletterswhicharecustomarilyissued
[emitti]inGreekheeitherdictateshimselforifdictatedinLatintrans-
latesintoGreek.Thus,inevidencefromvariousperiods,whileaclearfunctionalconnection
is made between incoming embassies and the letters which needed to be