Dear Mike,

Thank you for your replies. With the links you provided and the help of Wikipedia, I’ve realized that my understanding of the @era attribute was erroneous, and I believe I’ve found the information I was looking for.

As stated in Calendar (ICU4J 67.1), the recognized values for “calendar” are "buddhist", "chinese", "coptic", "ethiopic", "gregorian", "hebrew", "islamic", "islamic-civil", "japanese", and "roc". In a Java context ERA is a subclass of CALENDAR and accepts calendar-specific values. The only values I’ve seen in the links you provided are “AD” and “BC” for Gregorian and Julian calendars. I didn’t see the values specific to the Islamic calendar and other non-Christian calendars. However, most of these values can be found on Wikipedia. For the Islamic era, for example, we find this: In the West, this era is most commonly denoted as AH (Latin: Anno Hegirae) in parallel with the Christian (AD), Common (CE) and Jewish eras (AM) and can similarly be placed before or after the date. […] Years prior to AH 1 are reckoned in English as BH ("Before the Hijra"), which should follow the date.


Here is my previous example, corrected with this new information:


    <unittitle>Histoire de la conquête de l'Égypte, copie achevée le <date calendar="islamic" era="ah" normal="10660829" type="autre calendrier">27 Sa'ban 1066 de l'hégire</date></unittitle>

    <unitdate calendar="gregorian" era="ce" normal="1656">1656</unitdate>.



Best regards,

Florent Palluault


De : Encoded Archival Standards List [mailto:[log in to unmask]] De la part de Mike Ferrando
Envoyé : mardi 23 juin 2020 03:49
À : [log in to unmask]
Objet : Re: Dates in era/calendar other than ce/gregorian


Dear Palluault Florent,

I think we should use the terms found here (link below).

The attributes in the unitdate element directly correspond to variables (members) for the ICU4j java classes.

In any case, I hope this makes sense.


Mike Ferrando

IT Specialist

Library of Congress

[log in to unmask]




On Monday, June 22, 2020, 3:02:28 PM EDT, PALLUAULT Florent <[log in to unmask]> wrote:



Dear colleagues,


The website dedicated to EAD best practice in French libraries was moved to a different platform about a year ago. You can now find it at The original normative process dates back to 2008-2012 and work started some months ago on an update. French libraries use EAD 2002 and there are currently no plans to move to EAD3.


While working on the <date> element we wondered whether someone could provide a full example of use in an era and calendar other than Common era / Gregorian calendar. There aren't any records of such use in our dataset but we would like to illustrate the possibility, in case some collaborative projects require both the Gregorian calendar and another calendar.


Would the following be correct for a manuscript in Arabic, dating from 1656, with the scribe mentioning in the colophon that he finished his work on 27 Shaʿbān 1066?


    <unittitle>Histoire de la conquête de l'Égypte, copie achevée le <date era="ie" calendar="hijri" normal="10660829" type="autre calendrier">27 Sa'ban 1066 de l'hégire</date></unittitle>

    <unitdate era="ce" calendar="gregorian" normal="1656">1656</unitdate>.


In this example, I made 2 assumptions:

·         Islamic era abbreviated as “ie”

·         Islamic calendar indicated as “hijri”.

Is there a list of terms for eras and calendars and their use in the context of EAD?


Thanks in advance.



Florent Palluault

EAD in libraries working group (France)


Head of special collections
Poitiers public library
Médiathèque François-Mitterrand - 4 rue de l'Université - 86022 POITIERS - FRANCE
[log in to unmask]