likewise was always mobile: there were clerks, but no offices. No ancient office building and no ancient desk will ever be discovered. Strikingly, when
ancient administrative departments acquired a metaphorical name, it was not that of an unportable item of furniture, but the scribe's portable roll-satchel,
the scrinium. Administration revolved around people, not around places or buildings, and not, despite those mountains of papyrus, around documents.
The documents were stored in archive rooms, some of which are known archaeologically. But although papers were kept, there were no filing cabinets,
card indexes, reference numbers, registration forms. Collections of documents were made by pasting them together in chronological or-by no means as
often as convenience would dictate-in alphabetical order. The codex, the presentation of documents as a book, was occasionally used, but the cartulary, a
choice of really important documents for frequent reference, was unknown. Papers were preserved in archives, but it was well known that in most
conditions papyrus did not keep well. Why did these things not matter? Because retrieval of documents from the archive was not a particularly urgent
consideration in its formation.
The tax assessment notice, the letter from the commanding officer, the tax receipt, the birth registration were used only once, in the process of checking a
particular tax collection, or implementing a decision. Access to the document might be required a second time, but probably only a tiny fraction of all
documents was ever looked at twice. The consultation of a document was a serious matter: 'for which reason, pious and benevolent Caesar, order that I be
given a copy from your commentarii as your father intended' says a petitioner to Hadrian (ILS 338). Administrative processes were a favour, a privilege, a
wonder, which is why on documents like this, where only what does credit to the purchaser of the inscription appears, what seem to us to be banal details
of this kind are recorded in full. So this one actually preserves Hadrian's orders to his secretaries: 'Stasimus, Dapenis, publish the decision or opinion from
the recorded version (edite ex forma)'. Authentication was a serious problem, never entirely solved, which helped prevent reliance on documentary
authority. The sardonychus or imperial signet-ring gave its name to a Palatine department (see, e.g., ILS 1677), but there were often rumours that it had
fallen into unauthorized hands.
The Emperors used codes, but only rather simple ones. One of the principal reasons for the abuse of the public post system was that there was no reliable
way of ensuring that only a limited number of people possessed authentic licences to demand hospitality and service. Distribution was another problem. It
is very hard for us to imagine how difficult, despite the efficiency of communications, the systematic I exchange of documentary information was. A letter
of Trajan to Pliny making an important administrative point need never have been known in next door I Asia, let alone Germania Inferior. This is perhaps
one reason why Pliny's heirs actually published his correspondence. This difficulty no doubt helped to discourage the formation of any monolithic
imperial administrative structure.