Ottoman
Sephardic Genealogy: An Introduction
Page
2.
by Dr. David Sheby
Continued...
Conversions
tables for the first of the month in the Turkish-named Hejira calendar
to the Gregorian calendar and first of the month of the fiscal calendar
to Gregorian calendar can be found in:
Table
1.3: Programs and sources for Ottoman calendar conversions |
Conversion
|
Type
of Program
|
Site
/ Link
|
Gregorian
Dates -Hijri
Hijri
- Gregorian Dates
|
Internet
Site
|
|
Gregorian
- Julian - Hejira - Ottoman Fiscal
|
Downloadable
DOS program
"COMPUTAS"
|
Written
by Dr. Gerhard Behrens, at the MELA (Middle East Librarias Association)
website
|
Ottoman
religious- Ottoman civil/fiscal -Gregorian calendars
|
Monograph
|
Birken,
Andreas: Die Zeitrechnnung Handbuch der turkishchen Philatelie.
Teil I: Osmanisches Reich. Arbeitsgemeinschaft Osmanisches
Reich/Turkei, 1995, 40 pp. ISBN 3-931114-00-7. Bilingual (German
and English). Available from: James
Bendon Ltd., P. O. Box 56484, 3307,
Limassol, Cyprus. Contact info as of 11/99: E-mail: jbendon@attglobal.net.
|
For
those who cannot download the "Computas" program, a table of calendar
equiavalents is in the monograph:
Birken,
Andreas: Die Zeitrechnnung Handbuch der turkishchen Philatelie.
Teil I: Osmanisches Reich. Arbeitsgemeinschaft Osmanisches Reich/Turkei,
1995, 40 pp. ISBN 3-931114-00-7. Bilingual (German and English).
Birken's
text contains tables for the conversion of dates among the (a) Ottoman
religious, (b) Ottoman civil/fiscal, and (c) Gregorian calendars. In
addition, the Ottoman fiscal months' names are printed in osmanlica
(page 9). Birken's monograph is available from: James
Bendon Ltd., P. O. Box 56484, 3307, Limassol, Cyprus. Contact info
as of April, 2001: E-mail: books@JamesBendon.com.
1.2.4
"Reading" the Birth Certificate
1.2.4
"Reading" the Birth Certificate: Month/Date Figure 1.1 reveals where
specific information is located on the printed Ottoman birth certificate
from the late 19 th and early 20 th centuries. The task is to decipher
the handwritten entries which can be difficult to read. Dates are the
one of the easier pieces of information to decipher. Figure 1.10 is
a blow up of a section (i.e., top row, 4 th box from the right) of the
Ottoman birth certificate shown in Figure 1.1. Using Figure 1.7's list
of Arabic numerals the Ottoman year of 1320 is identified, as are the
numerals "2" and "8" which correspond to a date in a month (to be identified).
Figure
1.10. The year (1320) and day (28) of month for the Ottoman birth
certificate (Figure 1.1) are identified from the table of Arabic
numerals in Figure 1.7.
Identification of the month is more difficult, but can be successful
through detailed comparison of the date of birth box contents with each
of the Ottoman months in Figure 1.9. In Figure 1.11 the month "Kanun-i-sani"
is displayed twice so that it can easily be compared with different
portions of the text. The letter sequences "SA-NY" (written here left
to right, but written right to left in osmanlica as shown in Figure
1.11) are detected as is the sequence "K-A". No other month (Figure
1.9) has similar letter sequences. Hence the month is Kanun-i-sani.
The Ottoman date is 28 Kanun-i-san 1320.
Figure
1.11: The month (for the date of birth in the Ottoman birth certificate
of Figure 1.1) matches "Kanun-i-sani" (from Ottoman months in
Figure 1.9). The resultant date 28 Kanun-i-sani 1320 is equivalent
to 28 January 1905 (Julian) or 10 February 1905 (Gregorian).
Using
Birkin (1995) a table of equivalent dates in the Julian and Gregorian
calendars can be constructed (Figure 1.12). Kanun-i-san 1320 is equivalent
to 28 January 1905 (Julian) or 10 February 1905 (Gregorian).
Figure
1.12: Ottoman civil-Gregorian-Julian date equivalents for the first
28 days in Kanun-i-sani 1320 derived from Birken (1995, above).
1.3
Importance of Foreign Languages
The
Sephardim's Ladino-culture embedded within the Turkish-speaking Islamic
Ottoman Empire guarantees that genealogical-related source material
will be encountered that will written in a variety of foreign languages,
alphabets and script styles (cf. Table 1.4). Working with foreign languages
is not a optional task for a dedicated Sephardic genealogist, but a
necessity and a vehicle to obtain intimacy with a distant era.
Table
1.4: Languages and alphabets encountered during Sephardic genealogical
research |
Language
|
Alphabet
used for writing the language
|
Modern
Turkish
|
Latin
letters with modern Turkish orthographic symbols. Useful for reading
modern Turkish publications (such as academic publications on Ottoman-period
studies) and identifying sites of tombstones (see below). Note:
Turkish fonts are available at http://www.turkey.org/fonts/win311.htm |
Ottoman
Turkish (osmanlica)
|
Arabic
alphabet with 3 additions, from Persian, for sounds like letter
"p". Important for reading Ottoman birth certificates, passports,
and other government prepared material. Official forms will be printed
in a variety of lettering styles, and their entries will typically
be handwritten in a manner difficult to decipher. |
Hebrew
|
Hebrew
lettering: for reading tombstone & dedication inscriptions, and
rabbinical responsa. Hebrew lettering is often used as the alphabet
for transliterating Ladino text written in solitreo (or Rashi) into
modern Hebrew for use, for example, in dictionaries. |
Ladino
|
Rashi-style
lettering: a font used for Ladino-language newspapers, and other
printed material (books) |
Ladino
|
Cursive
script: solitreo: for
reading late 19th century handwritten correspondence or notations
on personal memorabilia |
Ladino
|
Variant
cursive scripts, close to but distinct from solitreo: used on kettubot
(marriage contracts) and community lists from mid 19th century and
earlier. |
French
|
- The
records of the Alliance Israélite Universelle (AIU).
- The
Sephardic journals (French) ETSI and (Belgian) Los Muestros
and other contemporary scholarly works (e.g., dictionaries)
dealing with Ladino and Sephardic studies
|
Serbian/Cyrillic
|
The Cyrillic alphabet is used for writing Bulgarian
and Serbian, the national languages for two successor states of
the Ottoman Empire (Bulgaria and Serbia--formerly "Yugoslavia) |
1.3.1
Modern Turkish Orthography
Familiarity
with five letter forms in modern Turkish (Figure 1.13), without any
ability to speak Turkish, can be helpful to a Sephardic genealogist.
After
the establishment of the Republic of Turkey, in 1923, Arabic was abandoned
as the alphabet for writing the Turkish language (1928). In its place,
the Latin alphabet was adapted, with additional letters, for use with
Turkish. The modern Turkish letters of most use to a Sephardic genealogist,
due to their unique orthographic symbols are (Figure 1.13):
1)
lower and upper case "c" with a "cedilla" (comma-shaped mark);
2)
lower and upper case "g" with the curved diacritical mark "breve"
3)
lower-case "i" without a dot;
4)
upper case "i" with a dot;
5)
lower and upper case "s" with a "cedilla".
Figure
1.13 Modern Turkish Alphabet
Modern
Turkish Alphabet
The
following two examples illustrate how elemental knowledge of modern
Turkish orthography can be useful to a SG.
1.3.2
Example 1: Location of a Grave Site
A
SG posted an inquiry, then a photograph, to a Sephardic mailing list.
It was an inscription on a relative's gravesite, showing that the epitaph
which was written in Ladino, in Latin letters. The SG inquired if the
tombstone inscription might identify the country of burial, i.e., where
the grave was located. The spelling containing the Latin letter "z".
Because of the presence of Ladino in Latin letters, it was speculated
that the grave was either in the United States or in a Latin American
country.
After
publication of a photograph of the entire gravesite area, a list-member
noticed that the family name was written using the capital I with a
dot over it: an orthographic (spelling) unique to modern Turkish as
shown in Figure 1.14. Knowledge that this spelling innovation occurred
in Turkey in 1928 (and that Turkey's international borders were fixed
by treaty in 1923) allowed the grave's location to be fixed within the
borders of the modern Republic of Turkey.
Figure
1.14 SALINAS family plot in unknown Turkish cemetery. Circa 1970.
(Source: Adatto/Alfassa/Benson/Yerushalmi family)
A.
Note the dot over the capital letter "I". This is an indication
of modern Turkish orthography (spelling) and the location of the
tombstone (burial site).
B.
Note that the surname SALINAS was spelled in all capital letters.
Continue
to
Page 3.