Approaches To Arabic Dialects A Collection Of

Articles Presented To Manfred Woidich On The

Occasion Of His Sixtieth Birthday Martine Haak
download
https://ebookbell.com/product/approaches-to-arabic-dialects-a-
collection-of-articles-presented-to-manfred-woidich-on-the-
occasion-of-his-sixtieth-birthday-martine-haak-2119054

Explore and download more ebooks at ebookbell.com

Here are some recommended products that we believe you will be
interested in. You can click the link to download.

African Arabic Approaches To Dialectology Mena Lafkioui Editor

https://ebookbell.com/product/african-arabic-approaches-to-
dialectology-mena-lafkioui-editor-50265546

African Arabic Approaches To Dialectology 1st Mena Lafkioui Ed

https://ebookbell.com/product/african-arabic-approaches-to-
dialectology-1st-mena-lafkioui-ed-4679908

Approaches To The Study Of Premodern Arabic Anthologies Nadia Maria El

Cheikh

https://ebookbell.com/product/approaches-to-the-study-of-premodern-
arabic-anthologies-nadia-maria-el-cheikh-36088320

Approaches To The History And Dialectology Of Arabic Papers In Honor

Of Pierre Larcher Lam Blg Manuel Sartori

https://ebookbell.com/product/approaches-to-the-history-and-
dialectology-of-arabic-papers-in-honor-of-pierre-larcher-lam-blg-
manuel-sartori-5758974
Counterterrorism In Saudi Arabia New Approaches To Radical Threats
Rachael M Rudolph

https://ebookbell.com/product/counterterrorism-in-saudi-arabia-new-
approaches-to-radical-threats-rachael-m-rudolph-50669908

Approaches To Language And Culture Svenja Vlkel Nico Nassenstein

https://ebookbell.com/product/approaches-to-language-and-culture-
svenja-vlkel-nico-nassenstein-45034526

Approaches To Metaphony In The Languages Of Italy Francesc

Torrestamarit Editor

https://ebookbell.com/product/approaches-to-metaphony-in-the-
languages-of-italy-francesc-torrestamarit-editor-49056716

Approaches To Phonological Complexity Franois Pellegrino Editor

https://ebookbell.com/product/approaches-to-phonological-complexity-
franois-pellegrino-editor-49056732

Approaches To The History Of The Interpretation Of The Qurn 1st

Edition Andrew Rippin

https://ebookbell.com/product/approaches-to-the-history-of-the-
interpretation-of-the-qurn-1st-edition-andrew-rippin-49114206
APPROACHES TO ARABIC DIALECTS
 STUDIES IN SEMITIC
LANGUAGES AND LINGUISTICS

EDITED BY

T. MURAOKA AND C.H.M. VERSTEEGH

VOLUME XXXVIII
APPROACHES TO ARABIC DIALECTS
 APPROACHES
TO ARABIC DIALECTS
A Collection of Articles presented to Manfred Woidich
on the Occasion of his Sixtieth Birthday

EDITED BY

MARTINE HAAK, RUDOLF DE JONG, KEES VERSTEEGH

BRILL
LEIDEN • BOSTON
2004
 This book is printed on acid-free paper.

Library of Congress Cataloging-in-Publication Data

Approaches to Arabic dialects : A collection of articles presented to Manfred Woidich on the

occasion of his sixtieth birthday / edited by Martine Haak, Rudolf de Jong, Kees Versteegh.
p. cm. – (Studies in Semitic languages and linguistics ; v. 38)
Includes bibliographical references and index.
Consists of articles in English, French, and German.
ISBN 90-04-13206-6
1. Arabic language–Dialects. I. Woidich, Manfred. II. Haak, Martine, 1959- III. Jong,
Rudolf Erik de. IV. Versteegh, C. H. M. V. Studies in Semitic languages and linguistics ;
38

PJ6709.A75 2003
492.7'7–dc21
2003050254

ISSN 0081-8461
ISBN 90 04 13206 6

© Copyright 2004 by Koninklijke Brill NV, Leiden, The Netherlands

All rights reserved. No part of this publication may be reproduced, translated, stored in
a retrieval system, or transmitted in any form or by any means, electronic,
mechanical, photocopying, recording or otherwise, without prior written
permission from the publisher.

Authorization to photocopy items for internal or personal

use is granted by Koninklijke Brill provided that
the appropriate fees are paid directly to The Copyright
Clearance Center, 222 Rosewood Drive, Suite 910
Danvers , MA 01923, USA.
Fees are subject to change.

printed in the netherlands

 CONTENTS

Preface ........................................................................................ vii

Bibliography Manfred Woidich ................................................ xi

The Arabic of Rabì 'a: A q6ltu dialect of Northwestern Iraq ............ 1

Farida Abu Haidar
Zum Verbmodifikator là- in nordmarokkanischen Dialekten ................ 13
Jordi Aguadé
Variability reproduced: A variationist view of the [µ]/[∂] opposition in
modern Arabic dialects ................................................................ 21
Enam Al-Wer
Die arabischen Dialekte von Jaffa und Umgebung ............................ 33
Werner Arnold
Von an-"Àß6r (al-Qaßr) nach Ìgni (Ìgli): Ein Vorbericht zu einigen
arabischen Dialekten der Provinz 6r-Ra“ìdìya (Marokko) ................ 47
Peter Behnstedt
Les parlers arabes nomades et sédentaires du Fezzàn, d’après William
et Philippe Marçais .................................................................... 67
Dominique Caubet
Quadriliteral verbs in the Arabic dialects of eastern Arabia ................ 97
Clive Holes
Technicalities and terminology of raiding and warfare as revealed by
Bedouin oral narratives ................................................................ 117
Bruce Ingham
Jüdisches, christliches und muslimisches Arabisch in Mosul ................ 135
Otto Jastrow
Characteristics of Bedouin dialects in Southern Sinai: Preliminary
observations ................................................................................ 151
Rudolf de Jong
Variation and change in Arabic urban vernaculars ............................ 177
Catherine Miller
Remarks on ideophones in Nigerian Arabic ........................................ 207
Jonathan Owens
Negations in the dialect of es-Sal†, Jordan ........................................ 221
Heikki Palva
vi 

Unmarked feminine nouns in modern Arabic dialects .......................... 237

Stephan Procházka
Relative-clause marking in Arabic dialects: A preliminary survey ........ 263
Jan Retsö
On verbal nouns in colloquial and literary Arabic ............................ 275
Judith Rosenhouse
The Arabic dialect of women in Meknes (Morocco): Gender linked
sound changes? ............................................................................ 291
Harry Stroomer
De la grammaticalisation de ‘comme’ (comparatif ) en arabe .............. 309
Catherine Taine-Cheikh
Deixis et focalisation: La particule ta en arabe de Yafi' (Yémen) ...... 329
Martine Vanhove
Pidginization and creolization revisited: The case of Arabic ................ 343
Kees Versteegh
Is there an Orientalist linguistics? .................................................... 359
Abderrahim Youssi
Colloquial Arabic in the 17th century: Yùsuf al-Ma©ribì’s Egyptian-
Arabic wordlist .......................................................................... 373
Elisabeth Zack

Index of Varieties of Arabic .................................................... 391

Index of Persons ........................................................................ 395
 PREFACE

To find colleagues and friends of Manfred Woidich willing to con-

tribute to the collection of articles now in hand was not the most
arduous of tasks. When we first started to approach colleagues on
the subject of a Festschrift for Manfred—this was right under his nose,
during the fourth AIDA conference in Marrakech in April of 2000—
their reactions were heartwarming and encouraging. In part, this
enthusiasm to contribute must have been due to the scholarly stand-
ing of Manfred Woidich. But perhaps even more so to his person-
ality—modest, ambitious only to work hard and preferably in the
background—which, almost as a matter of paradox, earned him a
central position in the field of Arabic dialectology as one of its lead-
ing figures.
The bibliography of Manfred Woidich’s publications bears testi-
mony to his achievements in the field. He has written extensively
on many aspects of Arabic dialects, from negative constructions to
folk tales, from diglossia to phonology. His didactic materials for the
study of Egyptian and Standard Arabic have become the preferred
teaching manual in many language courses. The dialect atlas of the
Egyptian dialects, which he published together with Peter Behnstedt,
already stands as a monument of research in Arabic dialectology.
Their joint introduction to dialect geography, due to appear in the
Handbuch der Orientalistik series at Brill’s, will no doubt become essen-
tial reading for anyone working in this field. The present collection
is an homage to his achievements.
As editors, we have deliberately chosen to include only articles on
topics in the field of Arabic dialectology rather than inviting all
friends and colleagues working in different fields. In the resulting
volume various approaches to Arabic dialectology are represented,
reflecting the central themes of Manfred Woidich’s research. The
value of his work lies in the combination of his deep respect for fine
detail—and the patience to strive for the full hundred percent while
collecting field-work data—with a broad interest in languages in gen-
eral, both on the theoretical and practical level. He keeps himself
completely with novel approaches in general linguistics, eager to
apply new insights to as yet unanalysed phenomena; he is a true
viii 

polyglot, in the sense of someone who loves language as well as

languages.
Although unassuming in his professional activities, Manfred will
always take pride in hosting a dinner party at his home, serving
pasta al dente with home-made pesto sauce, lavishly accompanied by
exquisite Frankish wines—or German beer for those who prefer—
all in the jovial atmosphere of Middle Eastern hospitality. All three
of us, in our different capacities, have had intensive contacts with
Manfred over the years: research, teaching, administration, the gen-
eral state of affairs in Dutch Oriental studies and other topics, but
the highlights of these contacts have always been the dinner parties.
And yes, there is always a lot of talk about Arabic dialects. He
never misses an opportunity to point out an especially interesting
Arabic expression, usually from one of his beloved Egyptian dialects.
With his encyclopaedic knowledge of these dialects he has acquired
quite a reputation as a latter-day Professor Higgins, even in Egypt.
He is known to have asked complete strangers in the Nile Valley or
the Delta whether their mother came from some unknown village
in Upper Egypt—and he would usually ‘guess’ right, leaving his
anonymous ‘victim’ and those present in a state of utter bewilder-
ment. He is a true master at finding a parallel between a word he
hears in any context with a word or expression from some Egyptian
dialect—a habit the Moroccan waiters in the restaurants he visited
during the Marrakech conference became well aware of.
There is always a sense of finality to a list of publications in a
Festschrift, which in this case would create quite the wrong impres-
sion. We have therefore ventured to add an item ‘to appear’, the
Cairene Egyptian grammar, which in many respects will sum up his
formidable knowledge about the dialect of Cairo; it may well have
appeared by the time this book is presented to him. As we write
these lines, we are aware that the occasion of his birthday itself may
not be very pleasant for him: an entire afternoon wasted with fes-
tivities, while he could have been at his computer, working on yet
another publication on the dialect of il-Ba“andi in the Dakhla oasis.
We sincerely hope that he will forgive us for this intrusion when he
has had a chance to read the articles collected in this volume, and
find the festive afternoon memorable at least.
One of Manfred’s first experiences with dialect differences must
have been when his classmates in Germany would tease him with
his ach-Laut whenever he had come to the blackboard to be heard
  ix

by the teacher, calling “Woidich, freu dich! ” [voidix froi dix]. Today,
on his sixtieth birthday, and now that he has been an inspiring
teacher and a leading scholar himself for a respectable number of
years, on behalf of all contributors to this volume, we should like to
make this our wish and add

kull6 sana w-inta †ayyib, ya Farìd!

Diemen—The Hague—Batenburg Martine Haak

January 2003 Rudolf de Jong
Kees Versteegh
This page intentionally left blank
 BIBLIOGRAPHY MANFRED WOIDICH

1969 Negation und negative Sätze im Ägyptisch-Arabischen. Diss., Universität

München.
1973–1974 “Die 3.sg.f. Perfekt im Dialekt von il-Bi'ràt”. Mélanges de l’Université
Saint-Joseph 48.355–372.
1973a Review of Fatma M. Mahgoub, A Linguistic Study of Cairene
Proverbs, Bloomington & The Hague: Indiana University Press,
1968. Zeitschrift der deutschen morgenländischen Gesellschaft 123.403–407.
1973b Review of Abdel Ghany A. Khalafallah, A Descriptive Grammar
of Saeìdi Egyptian Colloquial Arabic, The Hague & Paris: Mouton,
1969. Zeitschrift der deutschen morgenländischen Gesellschaft 123.401–403.
1974a “Ein arabischer Bauerndialekt aus dem südlichen Oberägypten”.
Zeitschrift der deutschen morgenländischen Gesellschaft 124.42–58.
1974b Review of Motie Ibrahim Hassan, In-nàs wil-malik: People and
King, Folk Tales in the Cairene Dialect in Roman Transcription,
Copenhagen: Copenhagen University Publications Fund, 1971.
Zeitschrift der deutschen morgenländischen Gesellschaft 124.146–149.
1975 “Zur Funktion des aktiven Partizips im Kairenisch-Arabischen”.
Zeitschrift der deutschen morgenländischen Gesellschaft 125.273–299.
1978 “Bemerkungen zu den arabischen Dialekten Mittelägyptens”.
Zeitschrift für arabische Linguistik 1.54–63.
1979 “Zum Dialekt von il-'Awàm‰a in der östlichen ”arqiyya (Ägypten).
I. Einleitung, grammatische Skizze und Volkskundliches”. Zeit-
schrift für arabische Linguistik 2.76–99.
1980a “Das Ägyptisch-Arabische”. Handbuch der arabischen Dialekte, ed.
by Wolfdietrich Fischer & Otto Jastrow, 207–242. Wiesbaden:
O. Harrassowitz.
1980b “illi als Konjunktion im Kairenischen”. Studien aus Arabistik und
Semitistik, Anton Spitaler zum siebzigsten Geburtstag von seinem Schülern
überreicht, ed. by Werner Diem & Stefan Wild, 224–238. Wies-
baden: O. Harrassowitz.
1980c “Zum Dialekt von il-'Awàm‰a in der östlichen ”arqiyya (Ägypten).
II. Texte und Glossar”. Zeitschrift für arabische Linguistik 4.31–60.
1980d (with Peter Behnstedt) “Zum Sprachatlas von Ägypten”. Zeitschrift
für arabische Linguistik 5.176–192.
1982a (with Peter Behnstedt) “Die ägyptischen Oasen: Ein dialekto-
logischer Vorbericht”. Zeitschrift für arabische Linguistik 8.39–71.
1982b Review of Stig T. Rasmussen, Ma˙mùd Taymùr, il-Ma˙ba"
‰aqam talattà“a‰: An annotated phonemic transcription, Copenhagen:
Copenhagen University Publications Fund, 1979. Zeitschrift für
arabische Linguistik 7.91–92.
xii   

1983a Review of Mokhtar Ahmed, Lehrbuch des Ägyptisch-Arabischen,

Wiesbaden: O. Harrassowitz, 1980. Zeitschrift für arabische Linguistik
11.92–95.
1983b (with Peter Behnstedt). Karte Ägypten—Arabische Dialekte. (= Tübinger
Atlas des Vorderen Orients (TAVO) A VIII 12.) Wiesbaden: L. Reichert.
1984a “Volksmedizinisches aus den Oasen”. Papyrus 11/12 (Cairo).
1984b Review of Charles Vial, L’égyptien, tel qu’on l’écrit: Glossaire établi d’après
un choix d’æuvres littéraires égyptiennes contemporaines, Cairo: Institut
Français d’Archéologie Orientale du Caire, 1983. Der Islam 61.355–
356.
1985a (with Peter Behnstedt). Die ägyptisch-arabischen Dialekte. I. Einleitung
und Anmerkungen zu den Karten. (= Beihefte zum Tübinger Atlas des Vorderen
Orients, Reihe B (Geisteswissenschaften), 50/1.) Wiesbaden: L. Reichert
Verlag. 129 pp.
1985b (with Peter Behnstedt) Die ägyptisch-arabischen Dialekte. II. Dialektatlas
von Ägypten. (= Beihefte zum Tübinger Atlas des Vorderen Orients, Reihe B
(Geisteswissenschaften), 50/2.) Wiesbaden: L. Reichert. 561 pp.
1985c Enige aspecten der hedendaagse tweetaligheid in Egypte. Inaugural lecture
at the University of Amsterdam, May 20, 1985.
1985d Übungsbuch zur arabischen Schriftsprache der Gegenwart. Wiesbaden:
L. Reichert. [Schlüssel 1986.]
1987a (with Peter Behnstedt). Die ägyptisch-arabischen Dialekte. I. Texte. Teil
1, Nildelta. (= Beihefte zum Tübinger Atlas des Vorderen Orients, Reihe B
(Geisteswissenschaften), 50/3.) Wiesbaden: L. Reichert. 311 pp.
1987b Review of Mohamed S. Anwar, Be and equational sentences in Egyptian
colloquial Arabic, Amsterdam: J. Benjamins, 1979. Mediterranean Language
Review 3.132–134.
1988 (with Peter Behnstedt). Die ägyptisch-arabischen Dialekte. III. Texte. Teil
2, Niltal und Oasen. (= Beihefte zum Tübinger Atlas des Vorderen Orients,
Reihe B (Geisteswissenschaften), 50/3.) Wiesbaden: L. Reichert. 450 pp.
1989a “Langform versus Kurzform: Die Kardinalzahlwörter von 3 bis 10
im Kairenischen”. Jerusalem Studies in Arabic and Islam 12.199–232.
1989b “illi ‘dass’, illi ‘weil’ und zayy illi ‘als, ob’: Zur Reinterpretation von
Relativsatzgefügen im Kairenischen”. Mediterranean Language Review
4/5.109–128.
1989c Review of Renate Malina, Zum schriftlichen Gebrauch des Kairinischen
Dialekts anhand ausgewählter Texte von Sa'daddìn Wahba, Berlin, 1987.
Der Islam 66:1.169–171.
1989d “Zur Bildung der Verbalstämme in den Ägyptisch-Arabischen Dia-
lekten: Der II. und der III. Stamm”. XXIII. Deutscher Orientalistentag,
Würzburg (1985), Ausgewählte Vorträge, ed. by Einar von Schuler, 200–
210. Stuttgart: F. Steiner.
1990a Ahlan wa Sahlan: Eine praktische Einführung in die Kairoer Umgangssprache.
Wiesbaden: L. Reichert. [Schlüssel 1991.]
1990b “Einige Aspekte der Diglossie im heutigen Ägypten”. Amsterdam
Middle Eastern Studies, ed. by Manfred Woidich, 99–130. Wiesbaden:
L. Reichert.
    xiii

1990c Materialien zur Kenntnis des Kairenisch-Arabischen. (= Leermiddelen voor

de studie van de Arabische Taal en Cultuur, uitgegeven door de Vakgroep
Arabisch van de Universiteit van Amsterdam Nr. 2.) Amsterdam:
IMNO. 280 pp.
1990d “/a/: Ein vernachlässigtes Kapitel aus der Morphophonologie
eines arabischen Dialektes”. Amsterdam Middle Eastern Studies, ed.
by Manfred Woidich, 131–159. Wiesbaden: L. Reichert.
1990e (ed.). Amsterdam Middle Eastern Studies. Wiesbaden: L. Reichert.
215 pp.
1991a “Zur Entwicklung der Konjunktion a˙san-la˙san im Kairenischen”.
Festgabe für Hans-Rudolf Singer zum 65. Geburtstag am 6. April 1990
überreicht von seinen Freunden und Kollegen, I, ed. by Martin Forstner,
175–193. Frankfurt am Main: P. Lang.
1991b “Die Formtypen des Zustandssatzes im Kairenischen”. Zeitschrift
für arabische Linguistik 23.66–98.
1991c “Short /a/ in Cairo Arabic morphophonology”. Semitic Studies
in Honor of Wolf Leslau on the Occasion of his Eighty-fifth Birthday,
November 14th, 1991, II, ed. by Alan S. Kaye, 1632–1651.
Wiesbaden: O. Harrassowitz.
1992a “Vorangestellte Demonstrativa im Kairenischen”. Jerusalem Studies
in Arabic and Islam 15.195–219.
1992b (with Jacob M. Landau). “The Baladiyyàt A˙mad ilFàr: A note
on a modern Egyptian manuscript text”. Manuscripts of the Middle
East 6.59–70.
1993a “Die Dialekte der ägyptischen Oasen: Westliches oder östliches
Arabisch?”. Zeitschrift für arabische Linguistik 25.340–359.
1993b (with Jacob Landau). Arabisches Volkstheater in Kairo im Jahre 1909:
A˙mad ilFàr und seine Schwänke. (= Bibliotheca Islamica, 38.) Stuttgart:
F. Steiner. 489 pp.
1993c Bibliographie zum Ägyptisch-Arabischen. Amsterdam: IMNO, Uni-
versiteit van Amsterdam. [1st ed., 1989.]
1994a “Cairo Arabic and the Egyptian dialects”. Actes des premières
journées internationales de dialectologie arabe de Paris, ed. by Dominique
Caubet & Martine Vanhove, 493–507. Paris: Institut National
des Langues et Cultures Orientales.
1994b (with Peter Behnstedt). Die ägyptisch-arabischen Dialekte. IV. Glossar
Arabisch-Deutsch. (= Beihefte zum Tübinger Atlas des Vorderen Orients,
Reihe B (Geisteswissenschaften), 50/4.) Wiesbaden: L. Reichert.
513 pp.
1995–1997 “Negation in the Egyptian Arabic dialect of the Dakhla Oasis:
The morphological reinterpretation of a phonological rule”.
Mediterranean Language Review 9.13–28.
1995a “Das Kairenische im 19. Jh.: Gedanken zum ˇan†àwì’s ‘Traité
de langue arabe vulgaire’”. Dialectologia Arabica: A Collection of
Articles in Honour of the Sixtieth Birthday of Professor H. Palva,
271–287. (= Studia Orientalia, 75.) Helsinki: Finnish Oriental
Society.
xiv   

1995b “Al-Sa'ìd: Dialects”. Encyclopaedia of Islam. New edition, VIII,

866b–867b.
1995c (with Rabha Heinen-Nasr). Kullu Tamàm: Inleiding tot de Egyptische
omgangstaal. Amsterdam: Bulaaq. 322 pp. [2 ed., 1996; 3rd ed.
1999; first appeared under the title of Zay6 ba'∂u: Inleiding tot
de Egyptische omgangstaal, Amsterdam: IMNO, 1993.]
1995d Review of Stephan Procházka, Präpositionen in den neuarabischen
Dialekten, Vienna, 1993. Orientalistische Literaturzeitung 90.58–65.
1996a “Some cases of grammaticalisation in Egyptian Arabic”. Proceedings
of the 2nd International Conference of l’Association internationale pour
la dialectologie arabe, held at Trinity Hall in the University of Cambridge,
10 –14 September 1995, ed. by Joe Cremona, Clive Holes &
Geoffrey Khan, 259–268. Cambridge: University Publications
Centre.
1996b “Rural dialects of Egyptian Arabic: An overview”. CEDEJ:
Egypte-Monde Arabe (Cairo) 27–28.325–354.
1997 “Upper Egyptian Arabic and dialect mixing in historical per-
spective”. Humanism, Culture, and Language in the Near East: Studies
in honor of Georg Krotkoff, ed. by A. Afsaruddin & A.H.M. Zahniser,
185–197. Winona Lake, IN: Eisenbrauns.
1998 “Aus den Erinnerungen eines Hundertjährigen: Ein Text im
Dialekt von Balà† in Ost-Dakhla/Ägypten”. Estudios de dialec-
tología norteafricana y andalusí 3.7–33.
1999a (with Peter Behnstedt). Die ägyptisch-arabischen Dialekte. V. Glossar
Deutsch-Arabisch. (= Beihefte zum Tübinger Atlas des Vorderen Orients,
Reihe B (Geisteswissenschaften), 50/5.) Wiesbaden: L. Reichert.
409 pp.
1999b Review of K. Mörth, Kardinalzahlwörter von ein bis zehn in den
neuarabischen Dialekten, Vienna, 1997. Bulletin of the School of Oriental
and African Studies 355–357.
2000 “The Arabic dialect of ilBa“andi at Dakhla oasis (Egypt)”.
Proceedings of the Third International Conference of AIDA (Association
Internatonale de Dialectologie Arabe) held in Malta, 29 March–2 April
1998, ed. by Manwel Mifsud, 145–150. Malta: Salesian Press.
2002 “Zum Dialekt von al-Qaßr in der Oase Dakhla (Ägypten)”.
‘Sprich doch mit deinen Knechten Aramäisch, wir verstehen es!’: 60.
Beiträge zur Semitistik. Festschrift für Otto Jastrow zum 60. Geburtstag,
ed. by Werner Arnold & Hartmut Bobzin, 821–840. Wiesbaden:
O. Harrassowitz.
2003 (with Jan Hoogland & Kees Versteegh). Woordenboek Nederlands-
Arabisch/Arabisch-Nederlands. 2 vols. Amsterdam: Bulaaq.
To appear Grammatik des Kairenischen. (= Porta Linguarum Orientalium, 000.).
Wiesbaden: O. Harrassowitz.
 THE ARABIC OF RABÌ'A:
6
A Q LTU DIALECT OF NORTH-WESTERN IRAQ

Farida Abu Haidar

London

1. Introduction

“Not too long ago, in the fifties and sixties, our knowledge of Egyptian
Arabic was practically limited to one dialect, i.e. the dialect of Cairo.
It was considered the Egyptian Arabic dialect per se” (Woidich
1994:493). This statement can apply to several other regions in the
Arab world where the dialect of the capital city is sometimes taken
to be representative of the speech of the country as a whole. In fact,
for a long time, whenever anyone spoke of Iraqi Arabic they usually
meant the dialect of Baghdad. Manfred Woidich has been instru-
mental in bringing to our notice varieties of Egyptian Arabic, besides
the dialect of Cairo. Similarly, Otto Jastrow (1978), covering another
corner of the Middle East where Arabic is the principal language,
has considerably broadened our knowledge of Iraqi Arabic by pro-
viding descriptions of some of the major q6ltu dialects of the region
and identifying several hitherto unknown ones. Jastrow collected valu-
able data in situ long before major population movement to and from
the q6ltu-speaking region of northern Iraq led to dialect loss and
dialect shift, particularly in small towns and villages.
Jastrow (1994:121–122) has divided the q6ltu dialects of northern
Iraq and Anatolia into six main groups: Mardin, Siirt, Diyarbakir,
Kurdistan, Tigris, and Euphrates. He has also subdivided the Tigris
dialects into three branches: Mosul, Tikrit, and the non-Muslim
dialects of central and southern Iraq. In this study I wish to survey
some points of phonology and morphology in the spoken Arabic of
Rabì'a, a dialect of the Mosul branch of the Tigris group.1

1
The present study is the first part of a project on the spoken Arabic of the
region of Jabal Sinjar in Iraq.
2   

Rabì'a is a small town in the province of Mosul in Iraq, just inside

the border with Syria. It is situated at approximately 100 km north-
west of the city of Mosul on the railway line connecting Mosul with
Aleppo.2 Rabì'a, up until the 1990s, was a fairly thriving town whose
predominantly Muslim, Arabic-speaking inhabitants maintained close
contact with neighbouring towns in Syria, and especially Tall Kushik,
only 5 km from Rabì'a, and the nearest town to it. After the Gulf
War and the introduction of border patrols, lack of freedom of move-
ment and constant political upheavals in northern Iraq forced most
of the young and able-bodied to leave Rabì'a.3 Many fled to south-
ern Turkey where they found refuge among the Arabic-speaking
communities of Anatolia. Others made the journey to western Europe
as asylum seekers.
The original inhabitants of Rabì'a are sedentary, and their livelihood
has depended for generations on agriculture and sheep-rearing. Over
the years the sedentary population was joined by people of bedouin
stock, mostly of ”ammari origin, who moved north in search of fer-
tile pastures for their flocks, and settled in Rabì'a and neighbouring
villages. Differences between those who are of sedentary origin and
others of bedouin provenance are reflected in their speech as well
as in their religious observance. Those of bedouin origin are Shiite
and speak g6l6t 4 Arabic, while the sedentary inhabitants of the town
itself are mostly Sunni, their speech being a variety of q6ltu Arabic.
The present study deals only with Rabì'a q6ltu Arabic (RA), q6ltu
varieties being the speech of a large number of the Arabic-speaking
Muslim and non-Muslim inhabitants of the province of Mosul.5 My
findings are based on five hours of recordings of the speech of seven
one-time inhabitants of Rabì'a, as well as on direct one-to-one inter-
views, carried out between October 2000 and August 2001. The two
women and five men, on whose speech this study is based, left Iraq

2
Right up until the Iran-Iraq war, trains along this line used to start their jour-
ney in Baghdad, their final destination being Istanbul.
3
Because of large-scale emigration and lack of up-to-date records, it is impossi-
ble to know the number of inhabitants that still live in Rabì'a.
4
Blanc (1964:30), who coined the terms g6l6t and q6ltu, transcribed them as gelet
and qeltu, where e stands for IPA 6. Some scholars, however, prefer to transcribe
them as gilit and qiltu.
5
Johnstone (1975:89) states that q6ltu dialects “until fairly recently were spoken
at least as far south as Baghdad”. One should add that for a long time the Arabic
spoken in the province of Mosul consisted only of q6ltu dialects, spoken along with
varieties of Neo-Aramaic.
    ì' 3

between 1995 and 1998. The two oldest are a married couple liv-
ing in Turkey. The husband was born in 1962 and his wife in 1963.
Among the other five are a younger married couple, born in 1971
and 1973, now living in Switzerland. The remaining three men, born
in 1974, 1978 and 1980, are at present in Germany. Four of the men
had been shepherds in Iraq. The fifth had worked on his family’s
small holding where they planted mainly wheat and barley. The two
women were housewives. Only two of the men and the older woman
had had elementary education, ranging from two to three years.6

2. Phonology

2.1 Consonants
Like all q6ltu dialects, RA is characterised by the retention of uvu-
lar plosive q and velar plosive k, allophones of g6l6t g and ‘ respec-
tively. Thus compare:
RA Baghdad
q: qarày6b garày6b ‘relatives’
halqadd halgadd ‘so much’
ts6wwaq tsawwag ‘to buy provisions, he bought provisions’
x6rqa x6rga ‘rag’
rqàq rgàg ‘thin, flaky bread’ (a speciality of northern Iraq)
k: klèb ‘làb ‘dogs’
b6ki b6‘a ‘to cry’
˙akayt ˙6‘èt ‘I spoke’
f6kk fa‘‘ ‘jaw’
samak s6ma‘ ‘fish’
g and ‘, however, are realized in some loanwords in RA, e.g., g6llàb6yya
‘tunic’ (< Egyptian Arabic, where g is the reflex of MSA [ ). g can
also be found in loans from Turkish where k of the donor language
becomes g in RA,7 as, for example:

6
In dialect studies in situ I invariably rely on data from informants belonging to
different generations. However, as I was unable to go to Rabì'a in person, my
choice of informants was limited to those I could find outside Iraq, who all hap-
pened to be under the age of forty. Because of population movement after the Gulf
War it has become increasingly difficult to study dialects in situ in some of the more
inaccessible regions of Iraq.
7
The following loanwords occur also in the Arabic of Mosul and Baghdad.
4   

RA Turkish
gòmlag gömlek ‘vest’
yalag yelek ‘waistcoat, sleeveless cardigan’
rang renk ‘colour’
k becomes g before b in the form akbar ‘bigger, larger’, realized in
RA as agbar.8
‘ occurs in loanwords which include the following: ‘ày ‘tea’, ‘m6ntu
‘cement’ and qa‘a© ‘smuggled, illegal’.
In a number of g6l6t and q6ltu dialects p can occur in some forms,
borrowed either from Turkish or European languages. Phoneme p,
however, does not occur as frequently in RA as it does in the Arabic
of Baghdad for example. An example of a form with p in RA is
paßapòr† ‘passport’. The following are a few examples where p has
been preserved in the dialects of Baghdad and Mosul, but not in
comparable RA forms:
RA Mosul Baghdad
ban†arùn pan†a©òn pan†arùn ‘trousers’
b6rdàyi pa©dàyi parda ‘curtain’
bìkàb pìkàp pìkàp ‘pick-up truck’
In a few borrowings from Western languages b of the donor lan-
guages is realized as p in g6l6t varieties and in the Mosul dialect also.
The two most frequently occurring examples of b > p are pàß ‘bus’
and pàys6k6l ‘bicycle’. In RA these forms are realized as bàß and
bàys6k6l respectively.
Interdentals are usually realized as dentals in RA, with ∆ > t,
≈ > d and µ > ∂:9
Mosul RA
m6∆6l m6t6l ‘like’
∆nèn tnèn ‘two’
hà≈a hàda ‘this’
≈ìb dìb ‘wolf ’
µa©abtùnu ∂arabtùnu ‘I hit him’
˙aµµ ˙a∂∂ ‘luck, fate’

8
This is the only example of k > g in my data.
9
Dentals for interdentals is a characteristic of the speech of Christian Iraqis.
Jastrow (1994:120) found this phenomenon also “in most Jewish dialects of Iraqi
Kurdistan, in the dialect of the Jews of Nusaybin/Qàm6“li . . . and in the Diyarbakir
group in Anatolia”. Jastrow (1994), moreover, points out that there are in the q6ltu-
speaking region of Southern Turkey, Northern Iraq and Northern Syria as many
as three different sound shifts involving interdentals.
    ì' 5

There were no instances of ∆ > s, ≈ > z or µ > Ω in the idiolects

surveyed here.
As far as MSA ∂ and Ω are concerned, most Iraqi speech vari-
eties do not distinguish between the two. In g6l6t varieties, and in
some q6ltu dialects, notably Mosul, the two MSA phonemes have the
same sound and are realized as a continuant, whereas in Christian
Baghdadi Arabic, for example, as well as in RA, they are realized
as a stop.10 This complete loss of phonemic distinction between MSA
∂ and Ω is not characteristic of Iraqi dialects alone, but obtains also
in some other varieties of Arabic (Beeston 1970:18, n. 1).
The r > © shift, a salient feature of the Tigris group as a whole
( Jastrow 1994:120), does not occur in RA. Thus compare:
Mosul RA
©6˙tu r6˙tu ‘I went’
©6bi rabi ‘he grew up’
a©nab arnab rabbit’
ka©©àt karràt ‘times, instances’
†à© †àr ‘it flew’
˙a[èy6© ˙a[èy6r ‘stones, pieces of rock’
An interesting phenomenon in RA is the shift n > l in the form
©alam ‘sheep’.11 Elsewhere in the region the MSA form ©anam has
been preserved.
As in a number of dialects of the Tigris group, s and ß in RA
become z and Ω in the contiguity of b, as in the forms zbù' ‘week’
( g6l6t sbù' ), zbèy6' ‘weeks’ and Ωbà'i ‘my finger’ ( g6l6t 6ßba'i ). s and ß
are retained where they are separated from b by a vowel, as in asèbì'
‘weeks’, the more formal plural of zbù', and aßàb6' ‘fingers’.
The alveolar affricate [ is realized as ≥ by some informants, due
perhaps to Syrian influence. The following forms were heard with
both [ and its Levantine allophone ≥:
[6bb and ≥6bb (pl. [bèb/≥bèb) ‘well, watering-hole’
˙à[èy6r and ˙a≥èy6r ‘pieces of rock’
ßà[ and ßà≥ ‘griddle (for baking bread)’

10
Transcribed as µ and ∂ respectively.
11
Procházka (2002) has found that the n > l shift occurs in the Arabic of south-
ern Turkey. He gives as an example Turkish dönüm > dlùm (pl. dlùmàt) ‘a unit of
measurement’.
6   

MSA © in ©asal ‘to wash’ is realized as x in RA, thus: xasìl ‘washing’,

xasal6t ‘she washed’, etc. x < © occurs also in the Arabic of Baghdad,
but not in the dialect of Mosul where MSA © is preserved.
The pharyngeal fricative ' is sometimes elided and followed by a
long vowel, as in the following examples:
mòr6f ‘I don’t know (< mà a'r6f or ma a'r6f )’
bàx6d à†i ‘I take and give (< bàx6d a'†i)’
ya'm6l mà wè˙6d ‘he works with someone’ (< ya'm6l ma'a wè˙6d ).
In this last example the first ' is not elided.
As in a number of q6ltu dialects, h is elided in the form 6“h6r
‘months’ when it is preceded by the cardinal numbers 3 to 10, e.g.
tlat 6“6r ‘three months’, s6tt 6“6r ‘six months’, etc. In g6l6t dialects h is
usually retained.

2.2 Vowels

2.2.1 Short vowels

The short vowels in RA are a and 6 in medial position and a, i and
u in final position.
a in medial position occurs in stressed and unstressed open sylla-
bles: waraqàyi ‘piece of paper’, bala˙a ‘date, palm-tree’, manàra ‘minaret’,
ba'6d ‘after’, ˙awìs ‘clothes’, labas6t ‘she wore’.
a occurs in closed syllables in a guttural or emphatic environment,
or where it is in the contiguity of r and a back vowel: ma'[ùn ‘(tomato)
paste’, karra ‘once’, ballù†a ‘acorn’, ˙ayyi ‘snake’, ˙abbi ‘grain’, ba††a
‘duck’, “qadd ‘how much?’.
a occurs in final position where the preceding consonant is a gut-
tural, an emphatic or r: karra ‘once’, x6rqa ‘rag’, s6yyàra ‘car’, ballù†a
‘acorn’, bala˙a ‘date, palm-tree’, †aq†aqa ‘noise’.
6 occurs in closed syllables: ˙6lwi ‘pretty [fem.sg.]’, w6˙di ‘one
[fem.sg.]’, n6[màyi ‘star’, [6bb ‘well, watering-hole’, [6wwa ‘below, under-
neath’, m6n[al ‘scythe’, s6yyàra ‘car’, n6zzèli r6˙˙àli ‘people constantly
on the move [lit. people who stay the night and then move on]’,
qab6l ‘before’, ba'6d ‘after’, ˙ar6b ‘war’.
There are, however, in my data two forms with medial a where
6 might have been expected. These are madrasi ‘school’ and fatti ‘broth
with pieces of bread’. This could be because both forms are loan-
words, the first from MSA and the second, a culinary speciality,
from outside Rabì'a.
    ì' 7

In vCCaC forms referring to colour, the initial vowel is 6 in non-

emphatic, non-guttural environments. Thus compare: 6swad ‘black’,
but a˙mar ‘red’, ax∂ar ‘green’, aΩraq ‘blue’, abya∂ ‘white’.
i occurs in word-final position following non-gutturals and non-
emphatics: sani ‘year’, n6[màyi ‘star’, ˙6lwi ‘pretty [fem.sg.]’, w6˙di ‘one
[fem.sg.]’, xàli ‘maternal aunt’, qaryi ‘village’.
u occurs in 1st. ps. sg. perfect verbs: q6ltu ‘I said’, “6ftu ‘I saw’,
sama'tu ‘I heard’, nasaytu ‘I forgot’.
u occurs also as a suffixed 3rd ps. sg. pronoun: sama'nànu ‘we heard
it’, 'aßaru ‘he squeezed it’, abùnu ‘his father’, baytu ‘his house’.
In the q6ltu dialects of the Tigris group the final vowel of the fem-
inine morpheme is not elided in the construct state. In g6l6t dialects,
on the other hand, the comparable vowel is invariably elided. Thus
compare:
q6ltu-Iraqi Baghdad Mosul RA
q6ßßa quß߆a q6ßß6tu q6ßß6tu ‘his story/tale’
˙adìqa ˙adìqta ˙adìq6tu ˙adìq6tu ‘his garden’
s6yyàra s6yyàrathum s6yyà©6t6m s6yyàr6t6n ‘their car’
Similarly, on the affixation of pronominal suffixes to the form màl
‘of, belonging to’ an anaptyctic vowel is added in RA and other
Tigris branch q6ltu dialects. No such vowel occurs in g6l6t Arabic
when the pronoun is in the singular. Thus compare:
Baghdad RA
màlti màl6ti ‘mine’
màltak màl6tak ‘yours [masc.sg.]’
màlt6‘ màl6tki ‘yours [fem.sg.]’
màlta màl6tu ‘his’
But màlatna màl6tna ‘ours’
màlatkum màl6tk6n ‘yours [comm.pl.]’
màlathum màl6t6n ‘theirs’
According to Johnstone (1975:91), “In very few dialects can there
occur a series of short open syllables”, and “in the dialect of Tikrit
the vowel of the post-initial syllable is elided”. Johnstone gives among
his examples katbat ‘she wrote’ and waqfu ‘they stood’. In RA, how-
ever, the short vowel of the second syllable is not elided, so that the
above two examples are usually katab6t ‘she wrote’ and waqafu ‘they
stood’. Compare also:
8   

Baghdad RA
s6m'aw sama'u ‘they heard’
˙6lbat ˙alab6t ‘she milked’
l6bsat labas6t ‘she got dressed’
Johnstone (1975:92) states that in Tikrit Arabic “there are no forms
in which a final consonant cluster occurs”. One should point out
here, that in most Iraqi Arabic dialects, be they of the q6ltu or g6l6t
group, an anaptyctic vowel occurs invariably between the two final
consonants of comparable MSA CvCC forms. Thus:
MSA Baghdad RA
bard bar6d bar6d ‘cold’
ba'd ba'ad ba'6d ‘after’
†abl †abul †ab6l ‘drum’
˙uzn ˙6z6n ˙az6n ‘sorrow’

2.2.2 Long vowels

The long vowels in RA are è, ì, à, ù, and occasionally ò, all occur-
ring in medial position.
à and è: A salient feature of all q6ltu dialects is the "imàla (à > è )
that distinguishes this group of dialects from g6l6t Iraqi Arabic. "Imàla
is a feature that is widespread in RA. Thus compare:
Baghdad RA
nà'6m nè'6m ‘soft’
nsà'6d nsè'6d ‘we help’
wà˙6d wè˙6d ‘one’
fawàt6˙ fawèt6˙ ‘wakes [pl. of fàt6˙a]’
˙amàya ˙amèyi ‘my brother-in-law’
"Imàla does not occur in post-guttural, post-emphatic environments
or in the contiguity of r with a back vowel. Thus: qàl ‘to say, he
said’, xàli ‘my maternal uncle’, nahàr ‘day’, naßàra ‘Christians’, ˙a††àbìn
‘wood gatherers’.
ì: Vowel ì occurs mainly in adjectives of the CaCìC type and plu-
rals of the CaCèCìC type, e.g. CaCìC: kabìr ‘big’, katìr ‘numerous’,
ba'ìd ‘far’, qarìb ‘near’, xafìf ‘light’, ˙awìs ‘clothes’.12

12
This last form is unusual. In the Mosul dialect and the Arabic of Christians
of Baghdad and Basra the word for ‘clothes’ is ˙wès.
    ì' 9

CaC√CìC: dakèkìn ‘shops’, ˙awèwìn ‘animals’, xawèrìf ‘sheep’.

ù: RA ù, as in abùyi ‘my father’, axùyi ‘my brother’, aqùl ‘I say’,
a“ùf ‘I see’, is sometimes the equivalent of ò in comparable forms in
the Mosul dialect. Thus compare:
Mosul RA
zbò' zbù' ‘week’
tannò© tannùr ‘earthenware oven [for baking bread]’
ballò† ballù† ‘acorn’
ma'[òn ma'[ùn ‘(tomato) paste’
malµòm mal∂ùm ‘dried okra threaded with string’
ò: Long ò is rare, but can occur sometimes in loanwords, e.g., gòm-
lak ‘vest’, b6z-zòr ‘by force’. I also found in my data two compound
forms with medial ò, mòr6f ‘I don’t know (< mà a'r6f )’ and lòx6r ‘the
other (< MSA al-àxar)’. Long medial ò can also be heard in a“lòn
‘how?’ although this forms seems to be in free variation with kayf.13

2.2.3 Diphthongs
Diphthongs aw and ay occur in RA where ò and è might be expected.
Thus compare:
Mosul RA
mòß6l mawß6l ‘Mosul’
mòt mawt ‘death’
dò“ak daw“ak ‘mattress’
hònìki hawnik ‘there’
y6[òn y6[awn ‘they come’
bèt bayt ‘house’
“èx “ayx ‘venerable old man’
baqètu baqaytu ‘I stayed’
˙akèna ˙akayna ‘we spoke, we talked’
According to Jastrow (1994:120) “the diphthongs ay and aw have
been preserved in many q6ltu dialects”.

13
The use of kayf is no doubt due to contact with Syrian Arabic where Syrian
kîf is adapted to the sound system of RA. See following paragraph on diphthongs.
10   

3. Morphology

3.1 Verbs
As in all q6ltu dialects, the 1st ps. sg. of the perfect in RA ends in
-tu, as, for example, r6˙tu ‘I went’, sama'tu ‘I heard’, t'6llamtu ‘I learnt’,
ts6wwaqtu ‘I bought provisions’, rabaytu ‘I grew up’, ˙alabtu ‘I milked’,
x6ttartu ‘I made yoghurt’, [ara“tu ‘I ground [lentils, cracked wheat,
etc.]’, 'a[antu ‘I kneaded [bread]’.
The imperfect in RA is frequently preceded by a modifier ka refer-
ring to a continuous past action, or by prefix b- denoting a habitual
action:
ka y6[awn ‘they used to come’
ka t6bß6†u ‘she used to spread it’
ka t6˙l6b 6n-na'[èt ‘she used to milk the ewes’
ka y6sqùh6n ‘they used to water them’
bàx6d à†i ‘I give and take [lit. I take and give]’
baqùm w aq'6d ‘I move about [lit. I get up and sit down]’
btàk6l t6“rab w tnèm ‘she eats, drinks and sleeps’
Of the two modifiers ka is the one that occurs more frequently.
Although it functions as a particle, ka is in all likelihood derived
from the auxiliary verb kàn. Prefix b- is less common, and could be
due to Syrian influence (cf. Behnstedt 1997; Arnold 1998). When
using more than one imperfect in the same sentence, informants
tended to use b- with the first verbal form only, as can be seen from
the last three examples.

3.2 Nouns
Like some q6ltu dialects, notably Christian Iraqi Arabic and the dialect
of Mosul, a number of feminine unit nouns have an ending in -àyi,
e.g., waraqàyi ‘a piece of paper’, n6[màyi ‘a star’, k6bbàyi ‘a patty of
crushed wheat and minced lamb’,14 bay∂àyi ‘an egg’.
Some feminine nouns have an ending in -6yyi. There are two
examples in my data: g6llàb6yyi ‘cloak’ and †annùr6yyi ‘skirt’.

14
This form is not to be confused with the Syro-Lebanese k6bbàyi ‘(drinking)
glass’. It should be pointed out that k6bbi (kubba in g6l6t dialects) is the regional spe-
ciality of the province of Mosul, its main ingredient being b6r©6l ‘crushed wheat’,
the staple diet of the region.
    ì' 11

The frequently occurring plural forms are of the CaC√CìC and

CaC√Ci type, where the second syllable is always long. CaC√CìC:
dakèkìn ‘shops’ (sg. d6kkèn), ˙awèwìn ‘animals’ (sg. ˙aywèn), xawèrìf ‘sheep’,15
masèkìn ‘poor people’ (sg. m6skìn), malè'ìn ‘scoundrels’ (sg. mal'ùn).
CaC√Ci: n6[èmi ‘stars’, †alèmi ‘(thick) loaves (of local bread)’, asèmi
‘names’, faqàri ‘poor (pl.)’, [a˙à“i ‘asses’.

3.3 Bound morpheme -6n

Jastrow (1994:120) states that there is “a very stable isogloss” in q6ltu
Arabic which separates the Anatolian branch from other q6ltu branches,
and that is the m/n ending of pronominal morphemes. Jastrow found
that in the Anatolian branch forms like bèthum (Dèr iz-Zòr), bayt6m
('Aqra) ‘their house’ and [ìtum (Dèr iz-Zòr), [ìt6m ('Aqra) ‘you [comm.pl.]
came’, for example, are bayt6n and [ìt6n respectively in Mardin. This
feature seems to be characteristic of RA also, as can be seen from
the following examples:
Mosul RA
y6sqùh6m y6sqùh6n ‘they water them, they give them to drink’
˙alab6t6m ˙alab6t6n ‘she milked them’
bèt6m bayt6n ‘their house’
màl6t6m màl6t6n ‘theirs’
b6nt6m b6nt6n ‘their daughter’

4. Conclusion

As can be seen from the above examples the Arabic spoken by the
sedentary inhabitants of Rabì'a is a variety of the Tigris branch of
q6ltu Arabic. RA has a number of similarities with the Arabic of the
city of Mosul, one of the better known dialects in the region. [For
this dialect see Jastrow’s contribution to this volume, Eds.] Where
RA differs considerably from the dialect of Mosul at the phonological
level, however, is primarily in the retention of MSA r, the allophone
of Mosul Arabic ©. At the morphological level, the main difference

15
I do not have the singular of xawèrîf in my data. Informants constantly referred
to ‘sheep’ as ©alam (sg. and pl.) or ©alamèt (pl.). For the singular they seemed to use
na'[i ‘ewe’.
12   

between RA and Mosul Arabic is the pronominal morpheme end-

ing in n instead of m. Modifier ka referring to a continuous past
action does not occur in Mosul where the auxiliary verb kèn (< MSA
kàna) is used instead. As far as verbal prefix b- is concerned, it is
difficult to ascertain from the idiolects surveyed here whether this is
a characteristic feature of RA or merely a loan from neighbouring
Syrian Arabic. Only a thorough investigation of a cross-section of
RA speakers of different ages and of both sexes would provide a
more in-depth knowledge of the dialect as a whole.

Bibliographical References

Arnold, Werner. 1998. Die arabischen Dialekte Antiochiens. Wiesbaden: O. Harrassowitz.

Beeston, A.F.L. 1970. The Arabic Language Today. London: Hutchinson University
Library.
Behnstedt, Peter. 1997. Sprachatlas von Syrien. Wiesbaden: O. Harrassowitz.
Jastrow, Otto. 1978. Die mesopotamisch-arabischen q6ltu-Dialekte. 2 vols. Wiesbaden:
F. Steiner.
—— 1989. Der arabische Dialekt der Juden von 'Aqra und Arbìl. Wiesbaden: O. Harrassowitz.
—— 1994. “The q6ltu dialects of Mesopotamian Arabic”. Actas del Congreso Internacional
sobre Interferencias Lingüísticas Arabo-Romances y Paralelos Extra-Iberos, ed. by Jordi
Aguadé, Federico Corriente & Marina Marugán, 119–123. Zaragoza.
Johnstone, Thomas Muir. 1975. “The spoken Arabic of Tikrît”. Annual of the Leeds
University Oriental Society 7.89–109.
Procházka, Stephan. 2002. Die arabischen Dialekte der Çukurova (Südtürkei). Wiesbaden:
O. Harrassowitz.
Woidich, Manfred. 1994. “Cairo Arabic and the Egyptian dialects”. Actes des pre-
mières journées internationales de dialectologie arabe de Paris, ed. by Dominique Caubet
& Martine Vanhove, 493–507. Paris: INALCO.
 ZUM VERBMODIFIKATOR LÀ-
IN NORDMAROKKANISCHEN DIALEKTEN

Jordi Aguadé
Universidad de Cádiz

Bekanntlich gebrauchen fast alle marokkanischen Dialekte einen

Verbmodifikator, um die allgemeine oder aktuelle Gegenwart auszu-
drücken.1 Am meisten verbreitet sind kà- und tà-, wobei kà- so ziem-
lich überall zu finden ist (Fes, Rabat, Casablanca, Tanger, Tetouan,
Larache usw.),2 inzwischen auch bei jüngeren Sprechern mancher
Dialekte, die traditionell andere Verbmodifikatoren kannten. Dagegen
wird tà- vor allem in den südlichen Dialekten verwendet (Aguadé
1996:198–199).3 Man findet aber auch andere Verbmodifikatoren wie
zum Beispiel dà- in dem jüdischen Dialekt von Debdou (Pellat 1952:
123, 133), qà-, südöstlich von Chefchaouen (Colin 1975:1195)), à-
bei den Branes nördlich von Taza und in Ouargha (Colin 1921:98)
oder ©à- bei den Írà©na (Guironnet 1954)).
In manchen prähilalischen Dialekten Nordmarokkos, zum Beispiel
in Anjra aber auch in anderen Gebieten von Jbala (¥bàla), kommt
ein Präverb là- vor (Vicente 2000:103–105).4 Dank der Untersuchung
von Vicente über den Dialekt von Anjra hat man sehr viele Beispiele
für den Gebrauch dieses Verbmodifikators.5

1
Eine Ausnahme bildet der jüdische Dialekt von Sefrou; vgl. Stillman (1981:239;
1988:40).
2
Einen Überblick über diese Verbmodifikatoren in Marokko sowie ihre Etymologien
findet man bei Aguadé (1996). Vgl. auch Durand (1991; 1994:89–90) und Grand’Henry
(1978:215–220).
3
Dazu auch Colin in de Prémare (1993–1999: II, 4).
4
Diese Arbeit ist besonders interessant, weil sie fast ausschliesslich auf den Aussagen
weiblicher Informanten beruht. Vgl. auch de Prémare (1993–1999: XI, 3) und
(1986, Texte von Z>hra einer Frau aus dem Dorf Ó>mmàra).
5
Der langsam auch von kà- verdrängt wird. Vicente (2000:103–104) bemerkt
dazu: “nos encontramos con una fuerte alternancia entre là- y kà-; siendo usado
el primero mayoritariamente por mujeres de edad avanzada y personas jóvenes
analfabetas, y el segundo por niños y niñas escolarizados y por la mayoría de los
hombres”.
14  

Besonders interessant ist der Dialekt von Chefchaouen (”6f“àw6n),

weil in dieser Mundart der Verbmodifikator flektiert wird (Natividad
1998:116–117). Wenn zum Beispiel das Verb m“a ‘gehen’ zusammen
mit dem Verbmodifikator konjugiert wird, so ergibt sich:

Sg. Pl.
3.m. là-yìm“i 3.c. là-yìm“ìw
3.f. dà-≈6m“i,
là-≈6m“i
2.c. dà-≈6m“i 2.c. dà-≈6m“ìw
1.c. nà-n6m“i 1.c. nà-n6m“ìw

Es gibt inzwischen eine beträchtliche Anzahl von Beispielen für den

Gebrauch dieser Verbmodifikatoren in Chefchaouen. In einem Aufsatz
mit Texten aus dieser Stadt, die von Natividad & Rahmouni (1996)
ediert wurden,6 findet man folgende Sätze: nà-n>""ìw †-†˙ìn ‘wir tun
Mehl hinein’, là-y≥ìbu s6††à“ d-6l-b>y†à∆ ‘sie bringen sechzehn Eier’, ma
dà-d6fh>m-“i ‘sie versteht nichts’, fìwàx là-≈6n'>s yìmm6ç ‘wann schläft
deine Mutter?’, 'là“, à-yìmma, dà-≈6xd>' hàd l>-'yàl ‘o Mutter, warum
täuschst du diese Kinder?’, là-y>'†ìha dàç5l ‘er gibt ihr etwas zu essen’.
Und aus einer noch unveröffentlichten Doktorarbeit von Moscoso
über den Dialekt von Chefchaouen stammen folgende Beispiele:7
h5wwa hna là-yìsk5n ‘er wohnt hier’, ˙>tta dà-d5wß>l n-6d-dචd-¶¶à≥6l
‘bis sie das Haus des Mannes erreicht’, àna nà-n>h∂>¶ ‘ich spreche
gerade’.
Der Verbmodifikator là- wird also in Chefchaouen folgendermas-
sen flektiert:

Sg. Pl.
3.m. là- 3.c. là-
3.f. dà-, là-8
2.c. dà- 2.c. dà-
1.c. nà- 1.c. nà-

6
Natividad & Rahmouni (1996). Die von den Verfasserinnen verwendete
Transkription wurde hier von mir geringfügig geändert.
7
Moscoso, El dialecto árabe de Chauen. Moscoso weist darauf hin, dass die Flexion
des Verbmodifikators charakteristisch für die Sprache der Frauen ist. Männer gebrau-
chen immer nur die unveränderliche Form là- oder sogar das gemeinmarokkani-
sche kà-.
8
dà- und là- werden frei variiert.
   LÀ- 15

Schon M. Cohen und Colin (1935:134) haben vor vielen Jahren die
Vermutung geäussert, dass der nordmarokkanische Verbmodifikator
là- aus dem Berberischen illa ‘er ist, es gibt’ entstanden sein könnte.9
Dieser Etymologie folgend, wies ich in einem 1996 erschienen Aufsatz
darauf hin, dass es noch dazu einen auffälligen Parallelismus zwi-
schen den flektierten Formen aus Chefchaouen und der Konjugation
des Verbs ‘sein’ im Ta“el˙it gäbe, wie aus der folgenden Tabelle
ersichtlich wird:

Sg. Chef. Ta“el˙it Pl. Chef. Ta“el˙it

3.m. là- illa ‘er ist’ 3.c. là- llan ‘sie sind’
3.f. dà- t6lla ‘sie ist’ (f. llant)
2.c. dà- t6llit ‘du bist’ 2.c. dà- t6llam ‘ihr seid’
f. t6llamt)
1.c. nà- lli© ‘ich bin’ 1.c. nà- n6lla ‘wir sind’

Da die Formen t6lla/t6llam > dà- als Sonorisierung t > d 10 und die
1.c. Sg. nà- als Analogiebildung zur 1.c. Pl. leicht zu erklären wären,
sprach ich mich ebenfalls für die Vermutung berberischer Substratein-
wirkung in diesem Fall aus, freilich ohne einen arabischen Ursprung
für diesen Vermodifikator ganz auszuschliessen (Aguadé 1996:205–206).
Nun, inzwischen halte ich diese Erklärung für ziemlich unwahrschein-
lich, denn es gibt sonst gar keine Belege dafür, dass arabische Dialekte
in Marokko bei Entlehnungen aus dem Berberischen auch die ber-
berische Flexion übernehmen.
Ausserdem gibt es eine viel einfachere Erklärung für die flektierten
Formen aus Chefchaouen: man kann sie nämlich ohne weiteres als
Ergebnis einer Assimilation des Verbmodifikators là- an die nachfol-
genden Präfixe der 2. und 3. Personen des Imperfekts deuten, also:
2. Sg. und Pl.: là- + d(6) . . . > dà-d(6) . . .
3. Sg. und Pl.: là- + n(6) . . . > nà-n(6) . . .
Somit wären diese flektierten Verbmodifikatoren aus Chefchaouen
lediglich Allomorphe des Präverbs là-: die Frage, ob dieser là- aus

9
Für Colin ist der Verbmodifikator là- “un résidu du berbère illà, 3è pers. masc.
sing. du verbe d’existence”. Auch Lévy (1990:550) und Tilmatine (1999) halten diese
Etymologie für gesichert.
10
In diesem Dialekt ist die Sonorisierung des Imperfektpräfixes t- ohnehin die
Regel: tàk5l > dàç5l ‘sie wird essen’, t6bdàw > d6bdàw ‘ihr werdet beginnen’, t6m“i
> d6m“i ‘du wirst gehen’ (Natividad & Rahmouni 1996:113–116).
16  

dem Berberischen illa stammt muss allerdings noch diskutiert

werden.
Für eine berberische Etymologie spricht schon manches. Wie bereits
vorhin erwähnt wurde, in nordmarokkanischen arabischen Dialekten—
es handelt sich dabei ausschliesslich um prähilalische Mundarten—
kommen die Verbmodifikatoren là-, dà-, à- und qà- vor. In manchen
berberischen Dialekten (im Rif und Mittlerem Atlas) werden die
Präverben la-, da-, a- und qa- verwendet, und zwar auch um die
aktuelle Gegenwart auszudrücken (Chaker 1984:972; Durand 1998:125;
Kossmann 2000:121–122)—der Parallelismus zwischen arabischen
und berberischen Dialekten ist in diesem Punkt auffällig. Die Subs-
trateinwirkung scheint hier also eindeutig, zumal die betreffenden ber-
berischen und arabischen Dialekte benachbart sind.11 Man bedenke
ausserdem, dass einen berberischen Ursprung bei diesen Präverben
zu vermuten ja naheliegend ist—schliesslich hat man es im Norden
Marokkos mit Dialekten zu tun, welche ohnehin einen starken Einfluss
des berberischen Substrats aufweisen.12
Nun, streng genommen darf man allerdings nur dann berberische
Substratwirkung annehmen, wenn die betreffende Erscheinung nicht
an einer anderen Stelle des arabischen Sprachgebiets vorkommt, das
dieses Substrat nicht aufweist.13 Und das ist eben hier nicht der Fall,
denn sowohl im Jemen als auch in Syrien—wo es bestimmt kein
berberisches Substrat gegeben hat—kommt ein Verbmodifikator là-
vor. Behnstedt (1985: Karte 83, Punkte 134, 140) erwähnt nämlich
in seinem Atlas der nordjemenitischen Dialekte einen Vermodifikator

11
Auch Tilmatine (1999:107–108) weist auf einen möglichen berberischen Ursprung
für die marokkanischen Verbmodifikatoren là- und qà- hin.
12
Eine übersichtliche Zusammenfassung dieser Merkmale findet man bei Diem
(1979:52–55). Neben manchen Erscheinungen, welche typisch für alle maghrebini-
schen Dialekte sind (etwa Reduktion der kurzen Vokale und Silbenbildungsregeln),
werden in den nordmarokkanischen Mundarten vor allem folgende durch Substra-
teinwirkung erklärt: Affrizierung von t, d, b; ∂ > †; ma ‘Wasser’ wird als Plural
behandelt; Veränderungen des syntaktischen Geschlechts einzelner Wörter; Genitiv-
konstruktionen des Typs xtu dd6-m˙6mmed ‘die Schwester Mu˙ammads’. In der letz-
ten Zeit hat sich Tilmatine (1999) mit dem berberischem Substrat in Algerien und
Marokko befasst und eine ausführliche Aufstellung möglicher Fälle von Substratwirkung
zusammengestellt. Zum Thema vgl. auch Aguadé & Vicente (1997:225–227, 234–236),
sowie Lévy (1996).
13
Diem (1979:16): “Die betreffende arabische Erscheinung darf nicht an einer
anderen Stelle des arabischen Sprachgebiets erscheinen, das dieses oder ein entspre-
chendes anderes Substrat nicht aufweist, sofern die Verbreitung der Erscheinung
durch Migration oder Wellenbewegung ausgeschlossen ist”.
   LÀ- 17

là- für die aktuelle und habituelle Handlung, der in der Gegend von
'A††ah, Marays und Juban gebraucht wird.14 Vanhove (1993:191) hat
là- auch in der Ortschaft Dhàla' (südlich der von Behnstedt unter-
suchten Dörfer) belegt. Als Beispiel führt sie an: là-ybki ‘er weint
gerade’. Zur Etymologie dieses Präverbs werden von den genannten
Autoren keine Angaben gemacht.
Was Syrien betrifft, so erwähnt Behnstedt (1997:325) in seinem
Sprachatlas den Gebrauch eines Verbmodifikators là- um die Zukunft
auszudrücken für die Gegend von Soukhne sowie für andere Ort-
schaften.15 Es sei hier noch hinzugefügt, dass es auch in anderen
orientalischen Dialekten Verbmodifikatoren gibt, welche eine durch-
aus plausible (und rein arabische) Etymologie für den nordmarokka-
nischen là- liefern können. Um die Zukunft auszudrücken gebraucht
man in manchen ägyptischen Dialekten (im Delta sowie in Oberägypten
um Gina, Luxor und Idfu) einen Verbmodifikator là˙ (Behnstedt &
Woidich 1985: Karten Nr. 224, 225), welcher wohl aus ràyi˙ ‘gehend’
entstanden ist.
Der nordmarokkanische Verbmodifikator là- kann genau so gut
aus rà˙ (rà˙ > là˙ > là-)16 wie in den oben erwähnten ägyptischen
Dialekten entstanden sein. Und man sollte nicht ausser acht lassen,
dass es im Hassaniyya—im Süden Marokkos und in Mauretanien—
auch einen Verbmodifikator là- (für das Futur) gibt, dessen Ursprung
man ebenfalls mit einer rein arabischen Etymologie (nämlich aus
dem Partizip làhi ) erklärt (Cohen 1924:274; 1963:113).
Vanhove hat in ihrem oben erwähnten Aufsatz auf das Vorkommen
eines Verbmodifikators là- in zwei so weit auseinanderliegenden
Ländern wie Jemen und Marokko hingewiesen und mit Recht die
berberische Etymologie für die nordafrikanische Partikel aus diesem
Grund in Frage gestellt: “à la lumière des données yéménites (pour
l’instant inexpliquées), l’origine berbère du préverbe marocain est
peut-être à discuter, notamment en raison de l’éloignement géo-
graphique des zones berbérophones et arabophones concernées, à

14
In anderen Gegenden findet man là- und là'- um die Zukunft auszudrücken
(vgl. Behnstedt 1985: Karte 8). Einen Verbmodifikator là findet man auch im neu-
aramäischen Dialekt der Juden von Rustaqa (vgl. Khan 2002:402), der aber damit
nichts zu tun hat.
15
Vgl. auch Behnstedt (1994:61).
16
Das Verb rà˙ ist in Marokko auch gebräuchlich, hat aber meistens die Bedeutung
‘[am Abend] zurückkehren, nach Hause zurückkehren’: vgl. de Prémare (1993–1999:
V, 231) und Harrell (1966:122).
18  

moins que l’histoire ne montre des liens privilégiés entre elles”

(Vanhove 1993:191 und Anm. 16).
Nun, Migrationen hat es schon gegeben aber immer vom Jemen
nach Nordafrika und der iberischen Halbinsel so dass manche jeme-
nitische Züge sich sogar im andalusischen Dialekt nachweisen lassen
(Corriente 1989): diese Einflüsse und die Tatsache, dass man là- auch
in Syrien und Ägypten findet spricht eindeutig gegen eine berberi-
sche Etymologie für den nordmarokkanischen Verbmodifikator.

Bibliographische Hinweise

Aguadé, Jordi. 1996. “Notas acerca de los preverbios del imperfectivo en árabe dia-
lectal magrebí”. Estudios de dialectología norteafricana y andalusí 1.197–213.
——, & Ángeles Vicente. 1997. “Un calco semántico del bereber en árabe dialec-
tal magrebí: El uso de la preposición 'la en el comparativo”. Estudios de dialectología
norteafricana y andalusí 2.225–240.
Behnstedt, Peter. 1985. Die nordjemenitischen Dialekte. I. Atlas. Wiesbaden: L. Reichert.
—— 1994. Der arabische Dialekt von Soukhne (Syrien). II. Phonologie, Morphologie Syntax.
III. Glossar. Wiesbaden: O. Harrassowitz.
—— 1997. Sprachatlas von Syrien. Kartenband. Wiesbaden: O. Harrassowitz.
——, & Manfred Woidich. 1985. Die ägyptisch-arabischen Dialekte. II. Dialektatlas von
Ägypten. Wiesbaden: L. Reichert.
Chaker, Salem. 1984. “Aspect”. Encyclopédie Berbère VII, 971–977.
Cohen, David. 1963. Le dialecte arabe ˙assànìya de Mauritanie (parler de la G6bla). Paris:
Klincksieck.
Cohen, Marcel. 1924. Le système verbal sémitique et l’expression du temps. Paris: Imprimerie
Nationale.
Colin, Georges S. 1921. “Notes sur le parler arabe du nord de la région de Taza”.
Bulletin de l’Institut Français d’Archéologie Orientale 18.33–119.
—— 1935. “L’opposition du réel et de l’éventuel en arabe marocain”. Bulletin de
la Societé de Linguistique de Paris 36.133–140.
—— 1975. “al-Mahgrib, 7: Aperçu linguistique”. Encyclopédie de l’Islam (Nouvelle Édi-
tion) V, 1193–1198.
Corriente, Federico. 1989. “South Arabian features in Andalusi Arabic”. Studia lin-
guistica et orientalia memoriae Haim Blanc dedicata, ed. by Paul Wexler, Alexander
Borg & Sasson Somekh, 94–103. Wiesbaden: O. Harrassowitz.
Diem, Werner. 1979. “Studien zur Frage des Substrats im Arabischen”. Der Islam
56.12–80.
Durand, Olivier. 1991. “I preverbi dell’imperfettivo in arabo dialettale”. Rivista degli
Studi Orientali 65.1–10.
—— 1994. Profilo di arabo marocchino: Varietà urbane centromeridionali. Rom: Università
degli Studi La Sapienza.
—— 1998. Lineamenti di lingua berbera: Varietà tamazight del Marocco centrale. Rom:
Università degli Studi La sapienza.
Grand’Henry, Jacques. 1976–1978. “La syntaxe du verbe en arabe parlé maghrébin”.
Le Muséon 89.457–475; 90.237–258; 439–456; 91.211–224.
Guironnet, M 1954. “©à- un outil linguistique polyvalent chez les Ulàd Ya'qub des
Srà©na”. Hespéris 41.425–433.
   LÀ- 19

Harrell, Richard. S. 1966. A Dictionary of Moroccan Arabic: Moroccan-English. Washington,

D.C.: Georgetown University Press.
Khan, Geoffrey. 2002. “The Neo-Aramaic dialect of the Jews of Rustaqa”. ‘Sprich
doch mit deinen Knechten aramäisch, wir verstehen es!’: 60. Beiträge zur Semitistik. Festschrift
für Otto Jastrow zum 60. Geburtstag, hrsg. von Werner Arnold & Hartmut Bobzin,
395–409. Wiesbaden: O. Harrassowitz.
Kossmann, Maarten. 2000. Esquisse grammaticale du rifain oriental. Paris & Louvain:
Editions Peeters.
Lévy, S. 1990. Parlers arabes des Juifs du Maroc: Particularités et emprunts. Thése de
Doctorat d’État, Université Paris VIII.
——. 1996. “Repères pour une histoire linguistique du Maroc”. Estudios de dialectología
norteafricana y andalusí 1.127–137.
Natividad, Emma. 1998. “Le dialecte de Chefchaouen”. Peuplement et arabisation au
Maghreb occidental: Dialectologie et histoire, hrsg. von Jordi Aguadé, Patrice Cressier
& Ángeles Vicente, 109–120. Madrid: Casa de Velázquez & Zaragoza: Universidad
de Zaragoza.
——, & Aicha Rahmouni. 1996. “Textos árabes de Chefchaouen: Transcripción,
traducción y glosario”. Estudios de dialectología norteafricana y andalusí 1.139–155.
Pellat, Charles.1952. “Nemrod et Abraham dans le parler arabe des Juifs de Debdou”.
Hespéris 39.121–129.
Prémare, Alfred-Louis. de. 1986. La tradition orale du Mejdûb: Récits et quatrains inédits.
Aix-en-Provence: Edisud.
—— 1993–1999. Dictionnaire arabe français. 12 Bde. Paris: L’Harmattan.
Stillman, Norman A. 1981. “Some notes on the Judeo-Arabic dialect of Sefrou”.
Studies in Judaism and Islam presented to Shelomo Dov Goitein, hrsg. von Shelomo Morag,
Issachar Ben-Ami & Norman A. Stillman, 230–251. Jerusalem: Hebrew University.
—— 1988. The language and culture of the Jews of Sefrou, Morocco: An ethnolinguistic study.
Manchester: University of Manchester.
Tilmatine, Mohamed. 1999 “Substrat et convergences: Le berbère et l’arabe nord-
africain”. Estudios de dialectología norteafricana y andalusí 4.99–119.
Vanhove, Martine. 1993. “Note sur le dialecte q6ltu de Dhala' (province de Lahej,
Yémen)”. MAS-GELLAS (N.S.) 5.175–199.
Vicente, Ángeles. 2000. El dialecto árabe de Anjra (norte de Marruecos): Estudio lingüístico
y textos. Zaragoza: Universidad de Zaragoza.
This page intentionally left blank
 VARIABILITY REPRODUCED:
A VARIATIONIST VIEW OF THE [˛π ]/[Î] OPPOSITION
IN MODERN ARABIC DIALECTS*

Enam al-Wer
University of Essex, UK

1. Introduction

In traditional and urban Arabic dialectology it is customary to refer

to the change from interdental sounds to stop sounds as a case of
phonological merger. Naturally, in the technical usage of the term
‘merger’ it is assumed that, at some point in the distant or recent
past, an earlier state of affairs existed when the features in question
were kept separate by the native speakers. Often we have evidence
for phonological mergers in neighbouring and/or related dialects
which did not undergo these processes, or in the existence of relic
forms which preserve the earlier distinction, or, where mergers are as
yet incomplete, evidence can be found in intergenerational differences,
and other social stratification patterns, which provide attestations in
apparent time of the pre-merger state. In the case of the plain inter-
dental Arabic sounds, the merger of [∆] and [≈] with [t] and [d],
respectively, is historically and empirically substantiated. However,
evidence that a phonological merger has occurred which collapsed
the lexical set containing modern [µ] with that containing modern
[∂] (in either direction) is simply unavailable. None of the spoken

* This article is a revised version of a paper originally presented at AIDA 4 in

Marrakech. I wish to thank Professor F. Corriente for his promptness in respond-
ing to my electronic queries. Thanks also to the editors of this volume for the sug-
gestions and insightful comments, many of which have been incorporated, while
some have fallen on deaf ears. Any remaining pitfalls are entirely my own. Two
of my colleagues at Essex deserve mentioning: Mike Jones for being the scholar he
is: inspiring and generous, and Andrew Spencer for the many hours of enjoyable
and fruitful discussions. To both of them I am truly grateful.
22  -

dialects maintains the distinction presumed in the written form of

the Standard variety, no spoken dialect has both sounds in its pho-
netic inventory, and no dialect, however isolated, ‘preserves’ vestigial
forms of the distinction. In this article, I debate this rather peculiar
situation, and I suggest an alternative analysis.
In relation to the occurrence of interdental sounds, the phonetic
inventories of modern Arabic dialects can be classified into two types:

• Type I dialects, which represent a system empty of interdental

sounds altogether, as in the majority of the urban varieties in the
Levant. These dialects are assumed to have undergone a systematic
sound change, which has resulted in interdental and stop merger,
as illustrated below:

Schema 1: Scenario of events in Type I dialects

Plain sounds Outcome Phonetic property of the outcome
/∆/, [∆] → [t] /t/ [t], contains lexical sets with
etymological /∆/ and /t/.
/≈/, [≈] → [d] /d/ [d], contains lexical sets with
etymological /≈/ and /d/.

Emphatic sounds
/µ/, [µ] → [∂] /∂/ [∂], contains lexical sets with
etymological /µ/ and /∂/.

The course of events envisaged for the plain interdental and stop
mergers as outlined in Schema 1 is empirically tenable, as can be
evidenced in the dialects which did not undergo the merger, and,
perhaps even more convincingly, in the dialects which show stable
variability in the use of plain interdental and stop, or progressive
tendency towards the merger. But, empirical evidence for the merger
of the emphatic sounds (in this case in favour of the stop sound) is
unavailable in Arabic dialects, since there are no native dialects which
maintain the distinction. Theoretically, however, a sound change
which would have shifted the emphatic interdental sound to a stop
sound (to parallel the changes in the plain interdentals) is quite plau-
sible, given that interdental sounds occur relatively rarely in human
languages, and a change from fricative interdental to stop is a widely
attested phenomenon. We can assume that the system here simply
eliminates ‘cumbersome’ sounds wholesale.
   23

• Type II dialects, which have the full set of interdentals, plain [∆]
and [≈], and emphatic [µ]. The traditional (indigenous) dialects of
Jordan and many of the rural and Bedouin Levantine varieties are
examples of such dialects. In these dialects, the plain interdentals
are phonemically distinguished from their stop counterparts [t] and
[d], respectively, but the emphatic interdental has no stop counter-
part. Schematically:

Schema 2: Scenario of events in Type II dialects

Plain sounds
/∆/ [∆], /t/ [t] No merger
/≈/ [≈], /d/ [d] No merger

Emphatic sounds outcome Phonetic property of the outcome

/∂/ → /µ/ /µ/ [µ], contains lexical sets with
etymological /∂/ and /µ/, i.e. merger
in favour of the interdental.

Schema 2 appears problematic, in that the change is incredibly selec-

tive (singles out one sound from a system), and predicts a develop-
ment in the ‘wrong’ direction: from stop to interdental fricative in
dialects which continue to maintain a phonemic distinction between
other stop and interdental sounds. A change from stop to fricative
is, of course, attested in human languages, but such a development
tends to be systematic, as exemplified by the systematic shift of Proto-
Indo-European voiceless stops to fricatives (Grimm’s Law). Original
Arabic ∂àd, however, was not a stop sound, but most probably an
emphatic lateral fricative, which lost lateralisation at some stage. This
issue and its implications will be further explored later in this article,
but for now let us start from the premise that in both cases above
the emphatic sounds had undergone different and/or additional his-
torical developments from those which had affected the plain sounds.
As I will show presently, these differences are reflected in the pat-
terns of variation and behaviour of the plain versus the emphatic
sounds in contemporary Arabic dialects.
24  -

2. Data and discussion

In 1987, I investigated variation and change in the use of the inter-

dental sounds in three Jordanian towns, and on the basis of this
research I reported that these sounds were undergoing a linguistic
change which merged them with their stop counterparts (details are
in Al-Wer 1991). I, like other researchers, who investigated these
variables in Jordan or in neighbouring communities, treated the
change as a systematic development. A closer look at the statistical
details in my data and those provided by other researchers reveals
that the plain interdentals show a slower rate of change to stops
than the emphatic interdental change to emphatic stop (see Abdel-
Jawad & Awwad [1989] on a sample of Jordanian speakers, Al-Khatib
[1988] on the city of Irbid in north Jordan, and Jassem [1993] on
a sample of refugees from the Golan Heights now resident in
Damascus; to my knowledge these are the only studies to date which
have included the interdentals as variables). In these studies, the
change (if these are indeed cases of change in progress) from [µ] to
[∂] is considerably faster than the changes from [∆] to [t] and [≈]
to [d]. The plain interdentals [∆] and [≈] show very similar rates. I
revisited the town of Sult in Jordan (one of the towns investigated
in 1987) ten years later to collect data for a separate research pro-
ject. The smaller sample of data I have from the 1997 study pro-
vides information in real time on the progression of the change from
interdental to stop. The data show that the change has progressed
dramatically; in particular, the change from [µ] to [∂] now shows
near completion in the speech of young women, while the frequency
of occurrence of /∆/: [t] and /≈/: [d] has roughly doubled (from
28% to 45%). In my opinion, the differences in the behaviour of
the plain sounds and the emphatic sounds may not be accidental.
Rather, they suggest that the parameters governing the changes from
[∆] and [≈] to [t] and [d] are different from the parameters gov-
erning the change from [µ] to [∂], which may be related to the fact
that the change which affects the plain interdentals is a merger (a
phonological change), but the change which affects the emphatic inter-
dental is a straightforward sound change from fricative to stop (a
phonetic change). In turn, this raises the question of the member-
ship of the emphatic interdental sound in this set, and implies that
plain and emphatic interdentals should be analysed separately (not
as a system).
   25

It is not clear whether there is any basis for expecting mergers,

as a type of sound change, to progress differently from regular sound
changes which do not involve phonological mergers, which would,
for instance, explain the varying rates of change presented above.
On the basis of empirical data, variationist sociolinguists have devel-
oped a number of models which show that there are several radi-
cally different mechanisms of merger, which operate at different rates
(see for instance Trudgill & Foxcroft 1978; Labov 1994, chapter 10).
The case of the plain interdental and stop merger in Arabic dialects
fits in with the model referred to by Trudgill and Foxcroft (1978)
as ‘merger by transfer’, which involves a gradual transfer of words
from one category to another. Labov (1994:323) maintains that merger
by transfer is the slowest, compared with ‘merger by approximation’
and ‘merger by expansion’.
Sociolinguistic stereotyping of the interdental sounds in Jordan pro-
vides further indication that the plain and emphatic sounds should
be treated separately. One observes, for instance, that [µ] is extremely
stigmatised in Jordan (and in the Levant in general). This sound is
often used to mimic, in ridicule, speakers of the dialects which have
it, but no such stereotypes are associated with the use of [∆] or [≈].
The issue of how and why positive or negative social values are
attached to certain sounds is complex, and the explanations pro-
posed are often circular. It is possible, however, that when a sound
change is nearing completion and the use of the departing feature,
here [µ], becomes restricted to a small minority of speakers, its
markedness increases, and the sound itself is then perceived as obso-
lete and rustic.
An intriguing case of plain and emphatic interdentals progressing
differently, and in this case dramatically so, is reported in de Jong’s
(2000) description of the Biyyàµiy dialect in northern Sinai. In this
dialect the plain interdentals have shifted to stop but the emphatic
interdental [µ] largely remains unaffected, a further indication that,
in the mechanisms of change, while the plain interdentals move sys-
tematically, the emphatic interdental does not necessarily follow suit.
It appears that in Biyyàµiy /µ/ is not involved in variation, apart
from the cases where [∂] is used invariably in K-forms (de Jong
2000:331–332). The stability of [µ] in comparison with [∆] and [≈]
in this dialect is itself an interesting, but, I think, different issue. De
Jong (2000:331–332, n. 8) cites examples of other Arabic dialects
“. . . in which interdentals have developed less predictably . . .”, namely
26  -

the dialects of Bahrain, Siirt, Àz6x, and B6˙zàni. I understand ‘less

predictably’ here to mean that the interdentals [∆], [≈], [µ] in these
dialects did not shift to stop counterparts, but to labiodental/stop/stop
in Ba˙àrna Arabic (i.e. the dialect of Bahrain), labiodental/labioden-
tal/labiodental in Siirt, and sibilant/sibilant/sibilant in Àz6x and
B6˙zàni. Less predictable these cases may be, but the developments
are perfectly in line with principles of simplification through elimi-
nation of marked sounds in favour of more frequent and/or less
marked sounds. In comparison however, the Biyyàµiyya case is baffling.
For no apparent linguistic benefit, it disregards the time-honoured
constraint of economy in language change. De Jong’s proposal that
external social factors such as group identity may account for the
anomaly of the Biyyàµiyya system is plausible, especially when one
considers that the name of the tribe itself contains the emphatic
interdental sound. Take the following example for an analogy. In
Amman, the sound [g] has largely lost ground to ["], but traditional
Jordanian names are immune to this transition. Speakers who other-
wise consistently use ["] maintain [g] in proper names such as 'ugla
and 'gàb.
A phenomenon which is readily observable in the Arabic dialects
which have the interdental sounds (Type II dialects) is that the speak-
ers of these dialects very often read the orthographic representation
of ∂àd as [µ], while the speakers whose native dialects do not include
the interdentals read the interdentals as stops. Take, for instance,
the following sets of data as examples of the performance of some
Arabic speakers in reading tasks, a domain in which speakers usu-
ally monitor their speech quite closely (not the sort of data one nor-
mally seeks, I dare say). Standard Arabic norms are highly prescribed,
and given that newsreaders, in particular, are expected to adhere to
these norms, the difficulties they exhibit with the interdental sounds
are probably ‘beyond their control’, so to speak.
In the fist set I recorded four Moroccan speakers, from the Moroccan
satellite TV station. Two of the speakers were delivering formal
speeches from a written text, and two newsreaders. The speakers in
this set read all instances of interdental sounds, both plain and
emphatic, as stops. A sample is listed in Table 1 below (the sounds
in question are in square brackets).
   27

Table 1: Examples from the Moroccan speakers

Correct reading Actual reading English gloss
li [≈] àlika li [d] àlika Hence
[∆] alà [∆] at 'a“ar [t] alà [t] at 'a“ar Thirteen
[∆] àbit [t] àbit Steady, Stable
"a [∆∆] àniya "a [tt] àniya The second
"ittixà [≈] "ittixà [d] The undertaking (of a task)
muna [µµ] amàt muna [∂∂] amàt Organisations
yanta [µ] ir yanta [∂] ir He is awaiting

The Moroccan ‘errors’ are readily explainable with reference to the

absence of interdental sounds from the native dialects of the speak-
ers (assuming they are speakers of such dialects). The behaviour of
these speakers is, in essence, similar to the behaviour of, say, German
learners of English. In another set of recordings I made, the speakers
came from Jordan, Tunisia, Iraq and Yemen (presumably speakers
of Type II dialects, which preserve the interdental sounds). Two
speakers of each locality were recorded from national satellite TV
stations, delivering formal speeches from written texts, or reading the
news bulletin. Here, the speakers make no ‘mistakes’ at all with the
plain interdentals, but they read ∂àd as (voiced emphatic interden-
tal) [µ]. This error was most consistent in the case of the Tunisian
speakers, who read all instances of /∂/ as [µ], followed by the
Yemenis’ score of 94% wrong, Iraqi’s 76%, and the Jordanians’ 37%.
A sample of the Tunisian data is listed in Table 2 below.

Table 2: A sample of the Tunisian reading of ∂àd

Correct reading Actual reading English gloss
mufàwa [∂] àt mufàwa [µ] àt Negotiations
"alwa [∂] ' "alwa [µ] ' The situation
"ay [∂] an "ay [µ] an Also
[∂] idd [µ] idd Against
"i [∂] àfatan "i [µ] àfatan Besides, in addition to
fa [∂] lan fa [µ] lan Apart from (the fact that . . .)
qa [∂] iyya qa [µ] iyya Issue
28  -

If a phonemic distinction between [∂] and [µ] ever existed, these

speakers clearly have no cognitive access to such a distinction (whereas
they have no problem in making the distinction between plain inter-
dental and plain stop). The data from the Jordanian speakers is par-
ticularly interesting. The modern urban dialects in Jordan are currently
in a state of variability with respect to the interdental sounds: [∆] is
used alongside [t], [≈] is used alongside [d], and [µ] (for historical
/µ/ and /∂/) is used alongside [∂] (for historical /µ/ and /∂/). The
attested variability represents changes in progress, i.e. these dialects
(originally Type II) are undergoing a change from interdental to stop
for both plain and emphatic sounds (i.e. are becoming Type I). But,
it is noticeable that the speakers have no problems in re-splitting /t/
into [t] and [∆], and /d/ into [d] and [≈]. The difficulty they show
is confined to the distinction between the emphatic variants. I attribute
this difference to the fact that in the case of the plain consonants,
the phonemic distinction is available in the linguistic experience of
the speakers, even if, sometimes, they do not make this distinction
phonetically, whereas in the case of the emphatic consonants, there
is no such evidence on which speakers can make a phonemic split
when required, e.g. speaking or even reading the Standard variety.

3. ∂àd versus µà" in history

If one examines the historical developments of /∂/ and /µ/ and

their distribution in the old Arabic dialects, the existence of a stage
where [∂] and [µ] were distinguished phonemically becomes precar-
ious. The thesis that the emphatic stop descends from Proto-Semitic
voiced lateral fricative, say [], is now widely accepted. To my mind,
Steiner (1976) and Corriente (1974) present a strong argument in
this direction. Corriente (1974:51), starting from the stage where the
emphatic voiced lateral fricative was phonemically distinguished from
/µ/, outlines a convincing course of events leading to the present
situation. These developments can be envisaged as follows:
• The system contains /µ/ [µ], and // []. Given the phonetic
complexity associated with the production (and the acquisition) of
the lateral fricative, it is reasonable to suggest that it lost its marked
phonetic properties, and changed into something else. The ques-
tion is: what could it have changed into?. The simple answer is
   29

that it would have changed into the nearest, less marked, and
acoustically similar sound. If it were to change into a stop by los-
ing lateralisation and creating a total closure, important acoustic
information, namely friction, would have been lost. At this point,
it is worth noting that the complexity of the sound is much more
likely to cause problems for the learners (the new acquirers, the
children) than for the adults, and therefore acoustic clues are vital
for a successful reproduction. The most likely development, there-
fore, was that the lateral fricative lost lateralisation, and the fric-
tion was reproduced through an interdental narrowing. This means
that its phonetic (and acoustic) properties became very similar to
or identical with [µ], and the outcome was a merger in favour of
[µ], which then contained the lexical sets with etymological lateral
fricative as well as those which already had the emphatic inter-
dental. Type II dialects stopped here.
But, Type I dialects continue as follows:
• [µ] (which is now the result of the merger of [] and [µ]), and
the plain interdentals [∆] and [≈] change into stops [∂], [t] and
[d], respectively. Notice that the outcome of this change is a merger
in the case of the plain interdentals (since the system already contained
/t/ and /d/), but the outcome in the case of the emphatic sound
is not a merger, since the system did not have a voiced stop
emphatic sound; the latter is rather a straightforward sound change
with no mergers involved.
What the historical developments outlined here show is that Type
II dialects have never had a stop counterpart to [µ] in the first place,
and that neither Type I nor Type II dialects have ever had the [µ]
versus [∂] opposition. If an opposition never existed, then the sound
change, to [∂], which has affected or is affecting [µ] in many Arabic
dialects is not a case of merger. Strictly speaking, however, in the
case of Type I dialects, the change from interdental to stop must
have gone through a stage of variability, whereby the old and the
new forms were used. These dialects may have contained [µ] and
[∂] variably at some stage, as variants (allophones) of the same
phoneme, but not contrastively, i.e. a very similar situation to that
found in some contemporary Jordanian dialects, as explained above.
30  -

4. The phonemic status of /∂/ versus /µ/ in Classical Arabic

Written Classical Arabic suggests a phonemic distinction between

/∂/ and /µ/, and at first sight, the phonetic descriptions by the
ancient grammarians are suggestive of the existence of this distinc-
tion. Nevertheless, it is surprising from this viewpoint that none of
the spoken dialects has maintained this contrast. The comments by
the ancient grammarians are unclear, in that although they mention
different pronunciations, for instance Sìbawayhi described a sound
which in modern terms is likely to be a voiced lateral fricative [],
and a separate sound [µ], it is possible, in my opinion, that what
the ancient grammarians are describing is in fact a situation of vari-
ation similar to that observed in the modern dialects, where both
sounds occur but not in the same dialects. Moreover, it is clear that
the modern dialects did not evolve directly from the Classical stan-
dard, in so far as standard forms of languages are typically a more
or less conscious amalgam of dialects. The question then is whether
the dialects from which both Classical Arabic and the colloquial vari-
eties developed had this contrast. Indeed, in the pre-Islamic texts,
different dialects had different variants, and there are numerous
examples of the alternation of the two sounds in the same words in
the early Islamic era (see El-Gindi 1982). El-Gindi (1982:426–428)
also names at least seven different collections of the Qur"àn (maßà˙if )
which either use ∂àd or µà" to represent both sounds throughout.
Interestingly, he mentions that in the Qur"àn, the word [©ali:µ] is a
perfect rhyme of ['ari:∂]. Clearly, these words can only rhyme if ∂àd
and µà" had an identical phonetic realisation.
A possible conclusion from these observations is that the contrast
found in Classical Arabic between [µ] and [∂] represents an artificial
split, which may have come about in a variety of ways, for instance,
through an attempt to codify this variation on a lexical basis with
the possibility that lexemes incorporated in the standard may have
come from different [∂] versus [µ] dialects. Those speakers in my
data who do not make the ‘correct’ distinctions in reading Classical
Arabic may not in fact be making mistakes, but are simply main-
taining, or reproducing, the same variability which was there between
these sounds in the classical period. Finally, Clive Holes, who is
hereby dearly thanked, has brought to my attention the following
comment by Kofler (1940–1942:96–97), which sums up the situa-
tion aptly:
   31

“. . . ferner auch, dass bereits im Altarabischen keine einheitliche Aus-

sprache des ∂ vorhanden war und das Bild der heutigen Dialekte schon
dort vorgezeichnet ist” [. . . furthermore, that already in Old Arabic
no uniform pronunciation of ∂ existed, and that the picture of today’s
dialects had already been sketched out].

Bibliographical References

Abdel-Jawad, Hassan and Mohammad Awwad. 1989. “Reflexes of Classical Arabic

interdentals: A study of historical sociolinguistics”. Beiträge aus Forschung und Anwendung
122.59–282.
Al-Wer, Enam. 1991. Phonological Variation in the Speech of Women from Three Urban
Areas in Jordan. Ph.D. thesis, University of Essex, UK.
Al-Khatib, Mahmoud. 1988. Sociolinguistic Change in an Expanding Urban Context: A case
study of Irbid City, Jordan. Ph.D. thesis, University of Durham, UK.
El-Gindi, Ahmad A. 1983. al-Laha[àt al-'arabiyya fì t-turà∆. Tripoli: ad-Dàr al-'Arabiyya
li-l-Kitàb.
Corriente, Federico. 1978. “Î-L doublets in Classical Arabic as evidence of the
process of de-lateralisation of ÎÂD and development of its standard reflex”. Journal
of Semitic Studies 23:1.50–55.
Jassem, Zaidan A. 1993. Impact of Arab-Israeli Wars on Language and Social Change in
the Arab World: The case of Syrian Arabic. Kuala Lampur: Pustaka Antara.
Jong, Rudolf E. de. 2000. A Grammar of the Bedouin Dialects of the Northern Sinai Littoral:
Bridging the linguistic gap between the Eastern and Western Arab world. Leiden: E.J. Brill.
Kofler, Hans. 1940–1942. “Reste altarabischer Dialekte”. Wiener Zeitschrift für die
Kunde des Morgenlandes 47.60–130.
Labov, William. 1994. Principles of Linguistic Change: Internal factors. Oxford: B. Blackwell.
Steiner, Richard. 1976. The Case for Fricative Laterals in Proto-Semitic. New Haven,
Conn.: American Oriental Society.
Trudgill, Peter & Nina Foxcroft. 1978. “On the sociolinguistics of vocalic mergers:
Transfer and approximation in East Anglia”. Sociolinguistic Patterns in British English,
ed. by Peter Trudgill, 69–79. London: Arnold.
This page intentionally left blank
 DIE ARABISCHEN DIALEKTE VON
JAFFA UND UMGEBUNG

Werner Arnold
Universität Heidelberg

1. Einleitung

Im Rahmen eines gemeinsamen Projekts zur Erforschung der ara-

bischen Dialekte Israels, habe ich während eines rund zweimonatigen
Feldforschungsaufenthaltes Daten zu den arabischen Dialekten in Jaffa
(Yàfa) und Umgebung gesammelt.1 Dabei konnte ich drei verschie-
dene Dialekttypen feststellen:
1. Die Stadtdialekte der Christen und Muslime, die in den drei
Städten Yàfa, Ramla (ir-Ramle) und Lod (il-Lidd) gesprochen wer-
den. Die Unterschiede im Dialekt der einzelnen Städte sind außer-
ordentlich gering. Auch zwischen den Dialekten der Christen und
Muslime gibt es kaum Unterschiede. Im Gegensatz zu den Dorf-
dialekten sind die Stadtdialekte sehr vital und werden von vielen
Tausend Menschen gesprochen. In Yàfa leben beispielsweise 20.000
Araber, davon ca. 5000 Christen.2

1
An dem Projekt sind außer mir noch Simon Hopkins und Aryeh Levin von
der Hebräischen Universität Jerusalem, Rafi Talmon und Aharon Geva-Kleinberger
(Universität Haifa) und Otto Jastrow (Universität Erlangen) beteiligt. Ich danke der
German Israeli Foundation für die Finanzierung meines Forschungsaufenthaltes in
Israel. Außerdem danke ich meinen Kollegen in Israel, die mich bei der Kontaktauf-
nahme mit meinen Informanten unterstützt haben, und zwar Gabi Rosenbaum und
Nàßir Baßal (Universität Tel-Aviv) sowie Rafi Talmon, Aharon Geva-Kleinberger
und Menachem Mor (Universität Haifa). Nicht zuletzt gilt mein Dank allen mei-
nen Informanten, die mir nicht nur bereitwillig Auskunft gegeben, sondern mich
auch freundlich aufgenommen und unterstützt haben. Dazu gehören der Vorsitzende
der samaritanischen Gemeine in Holon, Benyamim Tsedaka sowie Ilyàna Fanùs
und ¥òr≥ Íùßu (ir-Ramle), Merwad Munayyer (il-Lidd) und 'Ali Yatìm (Yàfa).
2
Nach der Eroberung der Städte Yàfa, il-Lidd und ir-Ramle duch die israeli-
sche Armee im Jahre 1948 mußten die meisten Bewohner die Städte verlassen. Auf
den Internetseiten www.palestineremembered.com findet man folgende Angaben:
Yàfa: von den 70.000 Bewohnern blieben 3650; ir-Ramle: von den 17.000 Bewohnern
34  

2. Die Dorfdialekte in der Umgebung der drei Städte sind heute

völlig verschwunden, da fast alle Bewohner im Jahre 1948 nach
Jordanien geflohen sind. Nur wenige Familien haben sich in den
Städten der Region niedergelassen. Die jüngeren Nachfahren haben
ihren Dorfdialekt aufgegeben und sprechen jetzt den Stadtdialekt.
Nur über 70 Jahre alte Sprecher haben noch eine Erinnerung an
den Dialekt. Von diesen habe ich jedoch nur in il-Lidd zwei Infor-
manten gefunden, die ursprünglich aus den Dörfern Íarafand und Kufr
'Àna stammen.3 Alle Bewohner dieser Dörfer wie auch aller ande-
ren Dörfer der Region waren Muslime.4 Viele Städter haben Vorfahren,
die aus den Dörfern zugewandert sind. Daher findet man bei den
Muslimen der Städte manchmal Formen, die mit den Dorfdialekten
übereinstimmen. Auf diese Weise sind in den Städten kleinere Dialek-
tunterschiede zwischen Christen und Muslimen entstanden.
3. Der Dialekt der Samaritaner gehört nicht zu den autochthonen
Dialekten der Region, da die Samaritaner in Holon ursprünglich aus
Nablus stammen, wo die andere Hälfte des kleinen Volkes bis heute
lebt. Zu Beginn des 20. Jahrhunderts haben sich die ersten Familien
in Yàfa niedergelassen. Seit 1954 leben alle Samaritaner Israels in
Holon, wo sie auch eine eigene Synagoge haben. Trotz der Heiligkeit
des Gerizim-Berges in Nablus verlassen viele Samaritaner die West-
Bank wegen der ständigen Unruhen und lassen sich im sicheren
Holon nieder. Inzwischen übertrifft die Zahl der Samaritaner in
Holon mit 330 Personen bereits die Einwohnerzahl in Nablus, wo
nur noch 309 Samaritaner leben.

blieben 400; il-Lidd: von den 19.000 Bewohnern blieben 1052. Seit dieser Zeit
wuchs die Bevölkerung wieder an. In Yàfa lebten im Jahre 1958 6500 Araber, im
Jahre 1965 bereits 10.000. Darunter sind natürlich viele Araber, die wegen der gro-
ßen Anziehungskraft Yàfas aus anderen Gebieten Israels, der Westbank und sogar
Ägyptens zugezogen sind.
3
Es handelt sich um 'Ali i“-”aràyi', geb. 1942 in Kufr 'Àna, 11 km östlich von
Yàfa und seine Frau aus Íarafand il-'Amàr, 5 km nordwestlich von ir-Ramle. Alle
Angaben zu den Dorfdialekten in diesem Artikel stammen von diesen beiden
Informanten.
4
Nur wenige Dörfer hatten winzige christliche Minderheiten. In Íarafand leb-
ten beispielsweise 40 Christen unter 1910 Muslimen, in Bèt Da[an 130 Christen
unter 3710 Muslimen. Nur der kleine Ort il-La†rùn mit seinen 190 Bewohnern war
überwiegend christlich (Quelle: www.palestineremembered.com).
Other documents randomly have
different content
the sign q with a Ben-Ik), refers to B, while signs 2 and 6 belonging to god D, who occurred in the
preceding Tonalamatl, should be noted.

The pair at the right on the other hand is seated, the woman apparently on the curved handle of a
vessel. The head-ornament and hieroglyph of the female figure prove that she is the serpent deity H,
while the male figure is the rare black deity M, whom we find again with his sign on page 43a for
example; he holds a bone in his hand. Hieroglyphs 9 and 13 agree. The lower part of these hieroglyphs
is the fist with the thumb unfolded, the sign at the top seeming to be merely an empty outline (Muluc?)
and thus, like 1 and 5 of the preceding group, they seem to refer to a sacrifice offered to the death-god.
10 and 14 are again, strange to say, like 2 and 6 of the preceding group, the sign of D. 11 is the
hieroglyph of H, who is represented below as feminine, and that 12 is a complement of 11 is proved by
the upper part of this uncommon hieroglyph, which corresponds to the object in H's hand, and which is
repeated on page 18a with the same figure; compare also page 8b. 15 is surely the hieroglyph of M,
who is pictured below, as in the Tro. 2a and 22*a where the same M appears with the same hieroglyph,
and to him belongs in 16 the sign r, which I am inclined to consider the week of 13 days, and which
here, as on 14c, ends a section of 13 days.

Page 15c.

III III 12 II 14 III

Lamat Ix
Ahau Cimi
Eb Ezanab
Kan Oc
Cib Ik.

This is a Tonalamatl of ten parts, the days are to be read in the following order:—Lamat, Ix, Ahau, Cimi,
etc.

There are two figures, A probably conceived as feminine and D with the same head-ornament as on
page 10; both hold in their hands a Kin = sun. Hieroglyphs 2 and 6 are also the Kin sign, while 1 and 5
have the closed eye of A, but differ in their secondary parts, the sign suggesting the south being a
suffix in 1 and a superfix in 5; 1, however, has an affix, while 5 has as a prefix a sign differing from the
affix in 1. 3 and 4 are the signs of A, 7 that of D, next to which in 8 one would expect to see an Ahau,
but instead of this there is again the sign of H (borrowed from page 15b?).

This seems to end the subject of coition; now, in natural course, follows the subject of pregnancy, to
which I believe the following Tonalamatl is exclusively devoted.

Page 16a.

Kan 21 31
Cib
Lamat
Ahau
Eb.

There are no red numerals, hence the Tonalamatl seems to apply to any one of the initial week days.

Two women are portrayed, both of whom are stretching a hand forward and upward. There are 8
hieroglyphs of which, however, the top row is almost entirely obliterated; 3 and 7 in the lower row are
just alike, being the usual sign for woman.

There is a decided contrast between the two figures, which might suggest barrenness and fruitfulness.
Observation of their physical differences would give us that idea. Furthermore, the first carries on her
back an unfamiliar head, perhaps A's, while the second has the Ahau, Imix and Kan signs, from which
plants seem to be sprouting. The first is represented in the fourth hieroglyph by the sign c, which is
closely allied to the death deities, while the second woman is denoted by hieroglyph 8 which is the sign
of the deity E, the grain-god.

Pages 16a—17a.

In the following I will group together all the pages from page 16-23 as follows:—First, I shall discuss the
top thirds, then the middle and lastly the lower thirds. The sense, however, often seems to require that
the first third should connect with the second, and the second with the third; but I find it impossible to
determine exactly the intended order.

On pages 16a-17a, we find for the first time in this manuscript not a Tonalamatl, but in its stead all the
twenty days arranged in four columns, each of which ends with one of the regents of the year:—

Men Ahau Chicchan Oc

Cib Imix Cimi Chuen
Caban Ik Manik Eb
Ezanab Akbal Lamat Ben
Cauac Kan Muluc Ix.

This seems to establish the fact that the day of its birth was of importance to a new-born child.

Between each column and the next there is a picture and above each picture four hieroglyphs, which,
however, are mostly destroyed, so that much of the meaning of this passage is lost to us.

The first is an old man walking, who beyond doubt is N, the Uayeyab god, with a staff in his hand and
the signs Imix and Kan on his back. He is looking upward and is also pointing upward with his right
hand. Of his hieroglyphs only enough of the fourth is visible to enable us to recognize in it the regular
sign of N, 5 Zac. The second picture is again an old man walking with a stick, he is baldheaded and
hence is probably also N, as on page 12c. His hieroglyph might be the fourth of those written above
him, the other three are entirely unrecognizable. He has a carrying-frame on his back, but it is uncertain
whether he is carrying anything upon it.

The third figure is a woman who is pointing upward with one hand and with the other holding the
bundle on her back, which I am unable to explain (does it refer to the 14th Uinal—the end of
pregnancy?) and from which rises an object resembling a flame. Her sign is in the fourth place and q is
in the third. 1 and 2 are not legible and perhaps may be supplemented by the third picture on page 19c.
Finally, the fourth figure is F, who is sitting and has a Cimi sign on his back. His monogram is the second
of the hieroglyphs above him, the third is very appropriately b and the other two are not very clear to
me.

The first two pictures might designate a male birth, the first indicating wealth and the second poverty,
the third might denote a female birth and the fourth a still birth. But who can positively assert this!

Pages 18a—19a.

VIII 12 VII 12 VI 9 II 10 XII 9 VIII

Ik
Ix
Cimi
Ezanab
Oc.
This is a Tonalamatl of five parts with 20 hieroglyphs, which unfortunately are so much injured that no
signs comprehending the whole can be distinguished.

There are five women in a sitting attitude.

The first woman corresponds exactly to the third figure on page 15b. She is sitting on a bench, the
same implement is in her hand and there is also a serpent on her head, for which reason she likewise
reminds us of H. The third hieroglyph is hers, and the 4th sign is an Ahau.

The second woman holds in her hand the Kin sign; above it is the Yax sign and above this a little cross
between two dots (the numeral 18?). Compare pages 18c, 19c and 27b, and in the second part, 46b
and 50c. I shall venture no opinion regarding the hieroglyphs.

The third woman with the copal pouch hanging from her neck has nothing in her hand. She is pointing
upward with her right hand. Her hair seems to be wound in the shape of an 8 in horizontal position and
above her is a sign denoting the union of two parts. The hieroglyphs are entirely destroyed. Does this
represent the birth of twins?

The eyes of the fourth woman are closed, she is pointing forward with her hand and there is a bird on
her head. Nothing is left of the hieroglyphs.

Finally, the fifth is distinguished by a large nose-peg, which, as on 12b, resembles a flower. Her hand is
extended forward. The fourth of the hieroglyphs above her is her sign. There is nothing to be said
regarding the three others. Are these five women engaged here in presenting their thankofferings and
prayers of thanksgiving for the birth which has taken place?

Pages 19a—21a.

XI 13 XI 13 XI 13 XI 13 XI 13 XI
Ahau
Chicchan
Oc
Men
Ahau.

Instead of Men the Manuscript has incorrectly Eb. Ahau in the fifth place is superfluous, since we have
here a Tonalamatl divided into four equal parts.

The hieroglyphs are so nearly obliterated that we can no longer distinguish a common sign. There were
in all six signs for the first picture, of which the first two are above the day-signs, while the figures from
the second to the fifth have only four signs each, as follows:—

1 2 5 7 8 11 12 15 16 19 20
3 4 6 9 10 13 14 17 18 21 22.

All that can be distinguished here is that the 4th and 13th have the same cross b and that 6 and 10
probably contain the same head.

Each of the five pictures contains a woman sitting. In the first representation she sits opposite a male
figure, who bends down to her with his bird-head, which we have already seen on page 13c. In the
other four pictures the woman is holding the figure of a god on her lap. I do not recognize the god in
the first picture on page 20. In the second and third pictures he is related to A or the Moan and the first
figure on page 21 may represent the god D. These can only be new-born children represented by the
gods under whose signs they were born. It should also be noted that the second woman on page 20
has a serpent on her head and the third a bird. The bird's head resembles that on page 16c.
Pages 21a—22a.

The Cimi and Eb of the second column have changed places in the Manuscript. Instead of the X there is
an erroneous 2 and there is no initial VII.

VII VII 3 X 2 XII 7 VI 9 II 3 V 2 VII

Oc Ahau
Cib Cimi
Ik Eb
Lamat Ezanab
Ix Kan.

We have here a Tonalamatl consisting of 10 × 26 days, and the 26 days are subdivided into six parts. I
have just assumed that the 2 is wrong and the initial VII is wanting over the first column, yet the 2
followed by the laterally elongated head q might here, perhaps, be explained in some manner as the
sign of the day VII Oc.

Apart from this sign which occupies an entirely exceptional position, we have here 24 hieroglyphs, i.e., 4
for each of the six groups.

The fourth sign in the first five groups is in each case a Chuen combined with the cross b and the suffix,
which seems to be a knife, and also with a numeral, which, however, is not recognizable in the first
group; in the second it is a 3, in the third a 7, in the fourth a 5 and in the fifth a 3. What can these
numbers mean? 3 + 7 + 5 + 3 = 18, and Chuen with the meaning of 20 (especially in the inscriptions)
would be 18 × 20 = 360.

In the fourth place of the sixth group there is a compound character, the main part of which (top, right)
seems to be the sign for the thirteenth month, Mac, and which may also, as we shall see on page 24,
denote the entire Tonalamatl. It is again compounded with a Chuen, an uplifted arm and a kind of
suffix, and hence might denote the end of a Tonalamatl.

The remaining 18 signs are in the main destroyed. In the second of the fourth group we recognize the
lock of hair denoting a woman, in the third of the second group the superfix suggesting the south,
which we find above the Cimi sign, for example on page 13b. Lastly, the other third signs are in the
third group Imix-Kan, in the fourth group the head q, in the fifth the bird c and in the sixth a Manik sign
with prefix and superfix resembling the sign i; in a few places (24, 39a, 53a, 56b, 58b, 61a, 61c, 68c)
the prefix might have the meaning of 20.

Since the intention was to close this section on the next page, the space had to be used as economically
as possible, and instead of the six pictures to be expected, there is only one and that is the first. It is a
woman in whom I observe nothing characteristic except that she has a kind of cloak, which has fallen
down over the lower part of her body, and who therefore remains unexplained.

Pages 22a—23a.

II II II II 2 IV 8 XII 7 VI 10 III 12 II
Men Cib Caban Ezanab
Chuen Eb Ben Ix
Manik Lamat Muluc Oc
Akbal Kan Chicchan Cimi
Cauac Ahau Imix Ik.

The Tonalamatl is no doubt to be read in this way after the correction of a few inaccuracies in the
Manuscript.
The 20 days, all of which occur again here as on pages 16a-17a, should be read from the right top to
the left bottom, since they form but one series.

As a matter of fact Ezanab is distant 19 days from the future Caban, but 39 days distant from the
desired weekday of the same name (see my "Erläuterungen," p. 24). Thus we have here a period of 20
× 39 days = 780, i.e., a three-fold Tonalamatl. The three Tonalamatls represented on the pages
between the preceding passage (pages 16a-17a), where all the 20 days appear, and this, are of three
different kinds (5 × 52, 4 × 65, and 10 × 26). This in itself is very remarkable. Furthermore a fourth
kind of Tonalamatl seems to be introduced here, which embraces, as it were, these three Tonalamatls.

The hieroglyphs, which are mostly destroyed, were arranged in groups of four for each subdivision, in
the following order:—

II II II II
1 2 5 9 13 17
3 4 6 10 14 18
7 11 15 19
8 12 16 20.

Of the above the third hieroglyph of each group, i.e., 7, 11, 15, 19 (probably also 3) is always the same
and is the sign of D, the moon and night-god. In detail we should expect to find five pictures here, but
owing to lack of space only the first of these is given. It represents a deity with a Kan sign in its hand
and a serpent on its head, who is probably E, and he is falling down here in exactly the same manner as
the four deities on page 15 at the beginning of this section.

Now, which were the other four deities? Signs 8, 12, 20 refer to A, H and C. 16 is the laterally elongated
head q, to which Seler is inclined to refer the day Men, and Schellhas an undetermined deity I. On
account of its frequency this sign must have besides a more general significance. In addition, however,
we have in 14 and 18 the signs of F and B. 6 is uncertain, 10 is probably C, and the top row is entirely
illegible. If to these deities is added the D repeated five times in the third row, it will be seen that all the
important gods are grouped together here on the last page of this section.

Pages 16b—17b.

I will now attempt (for it cannot be more than an attempt) to separate into three parts, according to
their contents, the middle and lowest thirds of pages 16 to 23. The first part, 16b to 18b and 16c to
20c, contains six Tonalamatls with pictures of women, each of whom carries on her back the figure or
symbol of a deity. This deity can hardly be any other than the one to which the horoscope of the child
especially refers.

The first of these Tonalamatls, on pages 16b-17b, runs as follows:—

Muluc 13 4 35 (or 20 15)

Imix
Ben
Chicchan
Caban.

The red numerals are wanting and were probably forgotten.

The hieroglyphs stand thus:—

1 2 5 6 9 13
3 4 7 8 10 14
11 15
 12 16.

Of these 3, 7, 11 and 15 are the sign for women, 2, 6, 10 and 14 are likewise all the same sign, which is
repeated in the same places on pages 17c to 18c. I do not understand its meaning; it may have
reference merely to the carrying-frame. Instead of the four women, whom we should expect to find
here, only the first two are portrayed. The first carries B, whose sign is the first hieroglyph, while the
fourth hieroglyph is the sign q.

The second woman carries A to whom hieroglyphs 5 and 8 refer. The third woman would have carried
D, which is plainly proved by hieroglyphs 9 and 12, and the fourth, F, as follows from sign 13 and
probably also from 16 (q).

Pages 17b—18b.

Eb 11 7 6 16 8 4.
Kan
Cib
Lamat
Ahau.

Here again there are no red numerals.

The 24 hieroglyphs of the six divisions stand thus:—

1 2 5 9 13 14 17 18 21 22
3 4 6 10 15 16 19 20 23 24.
7 11
8 12

Again, six women should be portrayed here, but there are only four; the second and third are wanting.
The signs for the women are given in 3, 7, 11, 15, 19 and 23, but in 15 and 19 the prefix is different
from that of the rest. As from here on the women repeatedly carry a bird, the signs for this are 2, 6, 10,
14, 18 and 22, which are the symbol of a rising bird, as in the sign of the 15th Uinal (Moan), which in
my opinion generally coincides with the 13th month of 28 days.

The women pictured here have nothing in their hands, which they hold stretched forward, as is usually
the case in this section. The first woman carries a vulture on her head. Compare 8a. In regard to it see
also Schellhas, "Göttergestalten," p. 31. The hieroglyph of the vulture, which we find repeated on page
17c, 24, 37b, 46, 50, 65, is here hieroglyph 1, usually regarded as the sign of the bat deity, and near it
in 4 is q.

The second woman would have carried the black deity L (hieroglyph 5), to which q is added in 8.

The third would have had the dog, i.e., the lightning dog, which we find in hieroglyph 9 and in the
month sign Kankin; an a is added to them in 12.

The fourth woman carries A, as is proved by his signs in 13 and 16.

The fifth carries nothing; according to the hieroglyphs 17 and 20 she ought to carry D.

Lastly the sixth carries the Moan as is proved by signs 21 and 24.

Pages 16c—17c.

Muluc 8 13 13 13 8 10
Ix
 Cauac
Kan
Muluc.

This is a Tonalamatl of 4 × 65 days. The Muluc at the bottom is, therefore, superfluous. I have been
obliged to correct the 12 in the last column of the Manuscript by changing it into a 10. The red
numerals are again wanting.

This passage admirably continues the one in the preceding Tonalamatl containing the women carrying
birds, and is also divided into six parts.

The hieroglyphs stand thus:—

1 2 5 6 9 10 13 17 21
3 4 7 8 11 12 14 18 22
15 19 23
16 20 24.

Signs 3, 7, 11, 14, 19 and 23 (14 and 15 have changed places) denote women. Of the six women only
the first three are here portrayed.

The first carries the Moan with which signs 1, 2 and 4 agree perfectly. The second and third carry two
birds, which may be parrots of a different species. They are very seldom represented elsewhere and
hence their hieroglyphs, 5 and 9, with the added determinative 10 are unfamiliar. In 8 and 12 the well-
known determinatives a and c are added.

Judging by sign 13 the fourth woman would have carried the same vulture, which we see in the middle
section of this page; 15 and 16 are again signs c and q.

The fifth woman would have carried an unknown bird of prey, the signs of which are 17 and 18, and
18=10; 20 is again q, but with a superfix different from that in 16.

Finally the sixth woman, like the third in 17b, seems to have carried the dog, as is proved by sign 21,
but in 22 the symbol of a bird is again added. This passage ends in 24 with the well-known Imix-Kan.

Pages 17c—18c.

IV 15 VI 33 XIII 4 IV
Ahau
Eb
Kan
Cib
Lamat.

Here we again find the regular red numerals (Roman in my transcription of the text), which were
wanting in the last three Tonalamatls. That they were not added until after the black script and
drawings were completed, is evident in several passages of our Manuscript and also in this one, where
they have been faintly indicated in black by the scribe (or corrector). The absence of red numbers in the
passages 17b-18b and 16c-17c is an evidence that I was right in proceeding directly from the former to
the latter.

Of the 12 hieroglyphs, 2, 6 and 10 have again the form which we found on pages 16b-17b, and which
seems to refer to a carrying-frame; compare, however, the explanation of pages 25-28 below. The
women themselves are designated by hieroglyphs 3, 8 and 12. The first woman carries the god A and
hieroglyphs 1 and 4 are his regular signs. The second woman has on her back a Kin sign, above that a
Yax, and this combination overtopped by a cross between two dots also forms hieroglyph 5; compare
the upper section of the same page. That this hieroglyph is nothing else than a designation of god D
follows from hieroglyph 7. Finally the fourth woman carries a figure, which has a Moan sign for a head
and to which hieroglyphs 9 and 11 certainly refer.

Pages 18c—19c.

XIII 32 VI 20 XIII
Ahau
Eb
Kan
Cib
Lamat.

The first woman carries the god A, who is denoted by hieroglyphs 4 and 1, though somewhat irregularly
by the latter. 2 is the carrying-frame and 3 the woman herself.

The second woman has again the Yax-Kin sign on her back as in the preceding Tonalamatl, and
hieroglyph 5 is also a combination of these signs, but here in 7 we find, not the sign of D, but that of E,
to which also the Imix-Kan in 8 corresponds. 6 is again the carrying-frame, though, as is also the case
in 2, more indistinctly drawn than in the earlier Tonalamatls.

Pages 19c—20c.

XIII 11 XI 11 IX 11 VII 10 IV 9 XIII

Ahau
Eb
Kan
Cib
Lamat.

This is a Tonalamatl divided into five parts, to which 20 hieroglyphs belong. The hieroglyphs are in the
following order:—

1 2 5 6 9 10 13 14 17 18
3 4 7 8 11 12 15 16 19 20.

At places 2, 7 (6 and 7 have changed places), 10, 14 and 18 we find again the sign which we think
means a carrying-frame, while signs 3, 6, 11, 15 and 19 are those of the five women.

The first carries a figure with a Moan head and agreeing with this is the second death-god F in
hieroglyph 1 and his determinative in 4.

The second woman, who is seated, carries the same object regarding which I am still uncertain, which
is carried by the standing woman on page 17a. This object is denoted by hieroglyph 5 (w). Its
determinative is probably 8. It may perhaps be a step in the right direction to point out that this sign
suggests the god K.

The third, like the first, has a figure with a Moan head, with which a female form of A in 12 and
hieroglyph 9 accord.

The fourth woman carries the maize deity E. 13 is his sign and the food hieroglyphs, Imix-Kan in 16,
agree with it.

The fifth woman seems to carry the somewhat indistinct form of D, if this may be inferred from the
Ahau of the 17th sign. 20 is the universal sign a.
This ends the six Tonalamatls, which are represented in what I have called the section of the burden-
bearing women. Five other Tonalamatls follow, which again suggest the idea of conception, which we
met once before on pages 13c-14c.

Page 19b.

X 29 XIII 23 X
Ik
Ix
Cimi
Ezanab
Oc

The most frequent sign in the five Tonalamatls, which I have grouped together, is the cross b, which
plays the most important part in all the Tonalamatls, excepting the third, which differs from the rest also
in other respects. It is essentially the sign for union, referring in the case of the stars to their
conjunction and here to sexual union.

In this Tonalamatl we see the cross in hieroglyphs 1 and 5, the sign for woman in 2 and 6, and their
determinatives in 3 and 7.

The first woman has a deity facing her who is devoid of all characteristic marks, and sign 4 is also
nothing but the universal a.

The second woman whose eyes are closed, sits facing A, whose hieroglyph is in 8.

Pages 19b—20b.

VI 28 VIII 24 VI
Cib
Lamat
Ahau
Eb
Kan.

The arrangement of this Tonalamatl is very similar to that of the preceding.

Hieroglyphs 1 and 5 are again the cross, and 2 and 6 the signs for woman.

The first picture is wanting; hieroglyph 3 with the number 7 as a prefix denotes a deity with whom I am
not familiar. The same sign is found on page 50, left, middle; in 4 the usual head q is added.

Beside the woman in the second group—not facing her—is the serpent deity H, again, as on pages 11c
and 12b, with the nose-peg resembling a flower. His sign is 7 to which in 8 the familiar Ahau is again
added.

Page 20b.

II 20 IX 19 II 13 II
Cauac
Chuen
Akbal
Men
Manik.
The hieroglyphs stand thus:—

1 2 5 9
3 4 6 10
7 11
8 12.

The subject now passes into the province of astronomy. This is already proved by sign 1, which
represents the clouds, between which the sun or moon is usually pictured; the sun is probably omitted
here merely owing to limited space. Sign 3 suggests the storm-god K (compare pages 7a and 47 left) to
which in 2 the Ahau might be appropriately added, inasmuch as it rules the year here under
consideration as on pages 25b to 26c. On account of the Ben-Ik sign I see in 4 one of the months of 28
days as a more exact determination of time. Below the Ben-Ik a head is represented with eyes
apparently closed, and this head is repeated in 6 and 10, though, probably for lack of space, without
the Ben-Ik. In each of the three places a sign is used as an affix which might readily be the year sign,
contracted laterally.

The two similar hieroglyphs 5 and 9, which have the following form, are especially worthy of
consideration:—

The part on the right recalls by its trisection the sign r, which I regard as the week of 13 days and, in
fact, the interval between the two hieroglyphs is 13 days. On the left is the inverted figure of a person
in a squatting attitude, the head surrounded by stars as on pages 57b and 58b and a sign on the back
which may be a suggestion of the sun-glyph. In this figure, which occurs also in the Tro-Cort. and in the
inscriptions, I see the planet Mercury and I believe that that planet's retrogression (which lasts 17-18
days) or disappearance into the light of the sun during this week, is the subject of this passage. 7 and 8
are the sign for D with the usual Ahau, and 11 and 12 are the hieroglyphs of the death-god A.

Instead of three pictures there is only one here, viz:—a woman with nose-peg, sitting on a mat and
apparently waiting for something. We also find figures sitting on mats elsewhere, for example on pages
7b and 68b.

Page 21b.

VII VII 7 I 7 VIII 7 II 5 VII

Oc Ahau
Cib Cimi
Ik Eb
Lamat Ezanab
Ix Kan.

This is also a Tonalamatl of 10 parts (10 × 26). The first column should be read first from top to bottom
and then the second. The days are exactly the same as on page 21a, and here too Cimi and Eb have
changed places.

The hieroglyphs run thus:—

1 5 6 9 13
2 7 8 10 14
3 11 15
4 12 16.
The signs forming the hieroglyphs into groups are, in addition to the cross in 2, 6, 10 and 14, the heads
in 1, 5, 9 and 13 with an Akbal sign (indistinct in 9) which, by the lock of hair in 5, 9 and 13, refer to a
woman. This lock of hair is replaced by a hand in 1.

Sign 3, with which m in 4 is associated as a determinative, shows that the first group ought to have a
picture of the black god L grouped with a female figure.

The second group is the only one with a picture. On the right there is a female figure, which, judging by
the headdress, we have already met on page 19a. Opposite her sits the dog which we saw on page 13c.
Here (in sign 7), as on page 13c, the hieroglyph of the dog is combined with a Cimi sign, and this
hieroglyph is repeated in 8 with the sign c, which is so closely allied to Cimi.

For the third group the god A should have been represented with the woman, as is proved by sign 11 so
peculiarly combined with r as a superfix. To this hieroglyph a is added, doubtless referring to the good
days, as if merely to fill space.

The hieroglyphs of the fourth group do not, I think, convey a clear idea as to which deity belongs here.
His sign is 15, which is compounded of Manik and Chuen with a superfix, nor does the Cimi added in 16
shed light on the subject. As for 15 we have already found it on page 13c with the prefixed 4, which I
find prefixed in this way in at least 12 different signs.

Pages 21c—22c.

Caban 5 21 16 10
Muluc
Imix
Ben
Chicchan.

This is a Tonalamatl of five parts in which the red numerals are wanting.

The hieroglyphs are in the following order:—

1 2 5 6 9 11 13
3 4 7 8 10 12 14
15
16.

Among these are hieroglyphs which are common to all the groups:—the cross in 1, 5 and 9 and the
woman in 3, 7 and 15. In 13 this cross is replaced by another sign, perhaps that for the year of 360
days, and in 12 the sign for woman is replaced by the universal a.

Each of the three pictures contains a woman facing a deity. I will consider first the second picture in
which H is the deity, as is proved by hieroglyph 6 to which an Imix is added in 8, with the uplifted arm
prefixed as in 10c and 13a.

Between the first and third pictures there is some confusion. The first is D, for while his type inclines
more to that of N, the other old god of the Maya Olympus, comparison with 23c clearly shows that D is
intended here. But the year-sign on his head also suggests in some measure the Uayeyab god N and
moreover this sign does not belong to D and only occurs again with him on page 23c. Further, there is
no hieroglyph at all for D and instead we find in 2, 5 Zac, the regular sign of N. Also sign 4 fits N better
than it does D. Furthermore this passage relates to the day Ik, which might very well be the last day of
the year.

On the other hand the third picture contains, unquestionably, the figure of N. I look for his sign in the
11th hieroglyph, which is the head of an old man with a prefixed 4, referring to the four different forms
of N in the Kan, Muluc, Ix and Cauac years. The Ahau in 12, however, does not fit N, but D.

This confusion can only be adjusted by transferring D from the first group to the third and also,
perhaps, the sign of the woman in 3, which applies to all the three groups, and by transferring to the
first group N and the 11th sign of the third group.

The fourth group has no picture. It should have, as hieroglyph 14 shows, the god F, who represents
death by violence in human sacrifice and the chase. The hieroglyph Cimi in the 16th place is a suitable
sign for this deity.

Pages 22c—23c.

II 10 XII 12 XI 9 VII 6 XIII 7 VII 8 II

Oc
Ik
Ix
Cimi
Ezanab.

The hieroglyphs are arranged in the following order:—

1 2 5 6 9 13 14 17 18 21 22
3 4 7 8 10 15 16 19 20 23 24.
11
12

This Tonalamatl, the fifth and last of this section, presents much that is irregular and puzzling.

It can hardly be said that there are comprehensive hieroglyphs here, forming the heading of the six
groups. The sign for woman occurs only in 2, 8 and 24, and the cross b only in 14 and 18, but it is
sufficient to make it clear that here, too, connection with a woman is the principal theme. Let us pass,
therefore, directly to the single groups.

The first group contains A and a woman. The god, however, is not facing the woman but sits beside her.
The Cimi sign in 1, the familiar c in 3 and the unknown sign in 4 (=6) hardly explain this particular
proceeding.

The second group contains two persons who sit facing each other, but the representation is so obscure
and peculiar that it is difficult to determine which is the male figure and which the female. The hair of
the person sitting on the right stands up in a manner not found elsewhere. It forms a figure similar to
that which is issuing from the mouth of the dog on pages 13c and 21b. The Cimi sign in 5 and the sign
c in 7 are familiar, but the infrequent 6=4 remains a puzzle.

Uncertainty regarding the third group is increased by the fact that there is no picture belonging to it.
The well-known signs, 10 (Cimi) and 12 (q) afford no explanation, nor does the head with the uplifted
arm in 11, which we find with the same hieroglyph on pages 8a and 36a. The most puzzling is the 9th
sign, which is composed of two crouching persons leaning back to back, and who also appear in the
astronomical sections of the Manuscript on page 68a, not merely in the form of a hieroglyph, but also
carried out in a picture. In my article on the Maya chronology published in the Zeitschrift für Ethnologie
of the year 1891, I attempted to explain this Janus picture as meaning the change of the year, but that
interpretation would make no sense here.

The fourth group contains the woman opposite D, who is clad in the gala mantle and has on his head a
bird and apparently the sign for a year, and is designated by the Ahau in 16, while Imix-Kan in 13, b in
14 and a in 15 are rather meaningless.
The fifth group represents the woman united with A, who is designated by the Cimi sign in 17. 18 with
its b and 19 with its q display little that is characteristic, r in 20, which I think is the sign for the week of
13 days, invites further study. The sixth picture, which is the last, is very peculiar; it represents three
women sitting side by side denoting perhaps the virgins who still remain. Sign 21 as Imix-Kan, 23 as a
and 24 as sign of femininity supply nothing in the way of explanation. As 6, 9 and 20 are the
characteristic signs in the preceding groups, so here the characteristic sign is 22—an open hand holding
the day Ben—which perhaps designates these virgins by referring to the house in which they are held
fast by the hand. Cf. Tro. 23* d.

Now of the entire woman section closing with page 23 only the two Tonalamatls on pages 22b-23b
remain. These Tonalamatls again display very many peculiarities and seem to be but loosely connected
with the five Tonalamatls last discussed.

Page 22b.

III 13 III 13 III 13 III 13 III

Akbal
Men
Manik
Cauac
Chuen.

This is a regular Tonalamatl, in which the 52 days are divided into four equal parts.

The hieroglyphs are in the following order:—

1 2 5 6 9 10 13
3 4 7 8 11 12 14
15
16.

An Ahau is added here as the 17th sign, which is very unusual.

We find elements here forming the hieroglyphs into groups in three different ways.

1. The signs 1, 5, 9 and 13 designate the four cardinal points as they so often stand together in this
Manuscript in the order of East, North, West and South, i.e., in the sequence of the annual and not of
the diurnal course of the sun.

2. The hieroglyphs 2, 6, 10 and 14 are all alike and are the head with the Akbal eye, which in 6 is
closed.

3. The three persons pictured here all carry a Kan sign in their hands, probably as the offering they
have received. Similarly we found the Kan sign held in the hand twice on page 16b.

The first picture is B; his sign is the third with the q in 4 as a determinative, which has above it a Ben-Ik
sign.

The second figure is a goddess with a serpent as head-ornament, though we find in the 7th sign, not
her hieroglyph, but merely the one generally used to denote a woman. 8 is the usual a, which in my
opinion is the sign for the good days, to which also the Kan sign refers in the hands of the three
personages.

The third picture is that of the sun-god G; his hieroglyph is the 11th, to which in 12 is added the sign q,
the sign for the bad days, with a superfix.
The fourth picture is wanting. According to the 15th hieroglyph it should be the maize deity E. My
theory that 16 is the sign for the week of 13 days is supported by the fact that the division into 4 × 13
days is the prevailing one.

Page 23b.

VIII 12 VII 12 VI 12 V 12 IV 12 III 5 VIII

Kan
Muluc
Ix
Cauac
Lamat.

This is a Tonalamatl of 4 × 65 days divided as evenly as possible into 5 × 12 + 5. The 5th day added
after the 16th must be a mistake (suggested by the 5th day of the last section) for it is usually the first
of the days, which is repeated superfluously.

The hieroglyphs are:—

1 2 7 11 15 19 23
3 4 8 12 16 20 24
5 6 9 13 17 21 25
10 14 18 22 26.

Contrary to practice the first section has six hieroglyphs, and the other five but four each.

As the characteristic hieroglyph we find in 1, 7, 11, 15, 19 and 23 a sign, the meaning of which is still
undetermined and which we shall meet again on page 60, where it may refer to darkness.

The groups have in common, furthermore, the head without an underjaw and the hair gathered up in a
tuft in 4, 10, 14, 22 and 25 (in 18 perhaps represented by q, the evil days). We shall find this sign on
pages 25, 28, 30-35, 42-44 and 65-69, repeated a number of times in many instances. I consider it the
sign for fast-days. It appears also in the Tro-Cort. Associated with this sign here as in other passages
are the four sacrifices derived from the animal kingdom:—a haunch of venison, a bird, an iguana and a
fish. The fish is beyond doubt denoted by 3, the mammal by 21 and the bird by 13, and I believe,
therefore, that the iguana with its spiny back is denoted by 9. We find the four animals, though in a
different order, also on pages 29b-30b, 30b-31b and 40c-41c, as well as in Cort. 3-6 and 8, for example.
They seem to have a certain reference also to the four cardinal points.

Only the first of the six groups has a picture (I?). This represents a woman with a serpent in her hair,
holding in her hand a dish containing a fish. The woman is denoted by the fifth hieroglyph and the fish
by the third. The 6th sign is an Ahau, which is not quite intelligible here. Sign 2=5 Zac is very
remarkable; it is the hieroglyph of the Uayeyab days and of their god N. If this Ahau refers, as it often
does, to the god D, it suggests the relation between D and N, which follows from page 21c.

According to the 8th sign, the second group might refer to the serpent deity H, and the 9th sign would
not improperly denote the iguana.

In the same way sign 12 in the third group probably denotes the storm-god K, with whom the bird in 13
accords very well.

In the fourth group both the animal and the sign of fasting, belonging to it, are wanting, while 16 and
17 as well as the unlucky day in 18 clearly refer to the death-deity A.

The fifth passage belongs, as sign 20 shows, to the maize deity E and to this is added the haunch of
venison in 21.
In the sixth group we recognize Imix-Kan, the sign for food derived from the vegetable kingdom. It
stands beside the grain-deity E of the fifth group. I do not understand the vulture-head in 26.

The five deities specified here may be compared with those on page 24, which are denoted by
hieroglyphs 21-25 of the second column, though the agreement is not perfect.

This ends the first great section of the Manuscript, in which Tonalamatls are represented in
uninterrupted succession. We come now to a page which stands quite alone, being the first which treats
of astronomy and which ends the front of the first part of the Manuscript.

Page 24.

In my article "Zur Entzifferung IV" I discussed this remarkable page in detail and in what follows I shall
conform to that treatise, though omitting many things which since then have become the established
possession of science, and shall endeavor to shed a still clearer light upon other points.

This page presents in brief the subject which is more fully treated of on the front of the second part of
the Manuscript (pages 46-60).

The first problem it presents is to find periods in which the solar year (365 days) is brought into accord
with the apparent Venus year (584 days). This takes place in a term of 2920 days = 8 × 365 = 5 × 584.
Sequent to this is the still higher aim of bringing the Tonalamatl (260) into harmony with this period,
which is accomplished in 37,960 days (= 146 × 260 = 104 × 365 = 65 × 584).

The revolution of the moon (28), the ritual year (364 = 28 × 13) and the apparent revolution of
Mercury (115) come in question as secondary matters.

I will now give an approximate reproduction of the page:—

Hieroglyphs.
1 17 29 151,840 113,880 75,920 37,960
2 18 30 (4 × 37,960) (3 × 37,960) (2 × 37,960) (13 × 2920)
3 19 31 I Ahau I Ahau I Ahau I Ahau
4 20 32 185,120 68,900 33,280 9100
5 21 33 I Ahau I Ahau I Ahau I Ahau
6 22 34 35,040 32,120 29,200 26,280
7 23 35 (12 × 2920) (11 × 2920) (10 × 2920) (9 × 2920)
8 24 36 VI Ahau XI Ahau III Ahau VIII Ahau
9 25 37 23,360 20,440 17,520 14,600
10 26 38 (8 × 2920) (7 × 2920) (6 × 2920) (5 × 2920)
11 27 39 XIII Ahau V Ahau X Ahau II Ahau
12 28 40
13 11,680 8,760 5,840 2920
14 (4 × 2920) (3 × 2920) (2 × 2920)
VII Ahau XII Ahau IV Ahau IX Ahau
15
16 2,200 1,366,560 1,364,360
IV Ahau I Ahau I Ahau
8 Cumhu 18 Kayab 18 Zip.

First let me observe that I have restored the four large numbers at the top, which are almost entirely
effaced, as follows:—

1 15 10 5
 1 16 10 5
1 6 16 8
14 0 0 0.
0

And furthermore, at the right, bottom, I have substituted the third month for the second of the
Manuscript, which proceeding will be justified later on.

The least difficult portion of the contents of this page is the first series consisting of 16 members, each
being a multiple of 2920. It begins with the date I Ahau (which is always concealed in these series),
regularly stops at the month day Ahau (since 2920 = 146 × 20), but necessarily advances in the week
days by 8 days each (since 2920 = 224 × 13 + 8), until 37,960 is reached, when the day I Ahau again
appears (since 37,960 = 146 × 260).

According to my method of filling in the numbers, the top row of the page consists only of multiples of
37,960.

On the other hand, the four numbers of the second row from the top are more difficult. They are, it is
true, all divisible without remainder by 260, but otherwise they seem to be without rule, and they give
one somewhat the impression of a subsidiary computation such as one might jot down on a slip of
paper in the course of some important mathematical work.

Nevertheless, the following remarkable results are obtained when the first and third and the second and
fourth numbers are combined by addition or subtraction:—

1) 185,120 + 33,280 = 218,400, which is just 600 years of 13 × 28 = 364 days, 280 Mars years of 780
days, 840 Tonalamatls of 260 days or 7800 months of 28 days.

2) 185,120 - 33,280 = 151,840, i.e., precisely the highest number of the top row, = 416 solar years of
365 days each or 260 Venus years of 584 days each, i.e., the product of the days of the Tonalamatl
multiplied by the Venus years. We shall again find the 151,840 on page 51, and Seler ("Quetzalcoatl and
Kukulcan," p. 400) finds this same period on a relief of Chichen Itza.

3) 68,900 + 9100 = 78,000, i.e., 100 Mars years or 300 Tonalamatls. The half of this number, or
39,000, we shall find again on pages 69-73 by computation; also the whole 78,000.

4) 68,900 - 9100 = 59,800, i.e., 520 Mercury years of 115 days, or 230 Tonalamatls, or five times the
period of 11,960 days, in which these two periods are united. By computation again we find the 59,800
on page 58. This period of 11,960 days is, however, to the period of 37,960 in the proportion of 23:73,
i.e., 23 × 520:73 × 520. 23 is the fifth part of the apparent Mercury year, as 73 is of the solar year.

Let us now turn to the numbers, which form the bottom of my transcription, but only the left hand
lower corner in the Manuscript. Here, in the latter, we find the following (with the correction already
mentioned of the second to the third month):—

2200 1,366,560 1,364,360

IV Ahau I Ahau I Ahau
8 Cumhu 18 Kayab 18 Zip.

The first thing to be done is to arrange and fill out these numbers to suit our purpose.

The 2200 is clearly nothing more than the difference between the two high numbers. We can therefore
dispense with it.

Further, we find by the usual computation, that the second number belongs to the first date and the
third to the second. Hence the number corresponding to the third date is wanting from lack of space.
This number can be calculated from that date; it is 1,352,400. It would suit this date equally well if the
number were higher or lower by 18,980 or a multiple of 18,980; but it will be seen directly that it
agrees with the other two numbers only at the value given above.

Now, if we add to this passage the years in which the dates must lie, they are in the case of the date on
the left, the year 9 Ix, in the case of the middle date, the year 3 Kan, and of that on the right hand, the
year 10 Kan.

Then if we arrange the three numbers with the dates and years belonging to them, according to the
value of the first, this part of the page will run as follows:—

1,352,400 1,364,360 1,366,560

I Ahau I Ahau IV Ahau
18 Zip 18 Kayab 8 Cumhu
10 Kan 3 Kan 9 Ix.

Let us now consider the properties of the three numbers individually.

1) 1,352,400 = 28 × 48,300 and = 115 × 11,760, hence it is divisible by the month days of the year of
364 days and by the Mercury year. At all events this is the least important of the three numbers.

2) 1,364,360. This looks as if it referred particularly to the moon and to Mercury; to the latter since it is
equal to 115 × 11,864, and to the former if we assume that the lunar revolution has been fixed at 29⅔
days, in which case this number is exactly equal to 46,000 such lunations. If this last number be again
divided by 115, the number of days required for a revolution of Mercury, the quotient is 400, which is a
round number in the vigesimal system and which was therefore denoted by a single word, by Bák in the
Maya (according to Stoll) and by Huna in the Cakchiquel (according to Seler). 1,364,360, therefore, is a
Huna of lunar revolutions multiplied by the number of days in the Mercury period. Later on we shall find
the lunar revolution fixed at 29⅔ days.

3) 1,366,560. This is the most comprehensive number of the entire Manuscript, for it is divisible into
each of the following periods:—Those of the Señores de la noche or Lords of the Cycle (9 × 151,840;
this is, however, the first number of the top row), the Tonalamatls (260 × 5256), the old official years
(360 × 3796), the solar years (365 × 3744), the Venus years (584 × 2340), the Mars years (780 ×
1752), the Venus-solar periods (2920 × 468), the solar year-Tonalamatls (18,980 × 72), the Venus,
solar, Tonalamatl periods (37,960 × 36), and the periods which are generally designated Ahau-Katuns
(113,880 × 12).

We have next to consider the intervals which elapse between the three dates.

1) From 1,352,400 to 1,364,360 is 11,960 days, which period we have already found once on this page
by computation. 11,960, however, is equal to 104 × 115 and 46 × 260, i.e., the Mercury revolution and
the Tonalamatl combined. 11,960 is again equal to 32 × 365 + 280, and from the year 10 Kan to 3 Kan
it is actually 32 years, and from the date 18 Zip to 18 Kayab it is, in fact, 280 days. The day I Ahau
must be common to both dates.

2) From 1,364,360 to 1,366,560 is 2200 days, as the Manuscript expressly states. 2200, however, is
equal to 8 × 260+120, and the distance from the day I Ahau to IV Ahau is in fact exactly 120 days.
Further 2200=6 × 365+10; from the year 3 Kan to 9 Ix it is 6 years and from the date 18 Kayab to 8
Cumhu it is 10 days.

3) From these two statements the third follows. The distance from 1,352,400 to 1,366,560 is 14,160.
This contains first the 14040, in which both the Tonalamatl and the old official year of 360 days meet,
and second 120, which is again the interval between I Ahau and IV Ahau. But 14,160 is also equal to 38
× 365 + 290, and the interval between 10 Kan and 9 Ix is of course 38 years, and from 18 Zip to 8
Cumhu it is 290 days.
The numbers with which we have had to do here will again occupy our attention further on, especially
the 2920 and the 37,960 on pages 46-50, the 11,960 and 115 on pages 51-58, and the 14,040 on page
73.

That these computations are not confined to the Dresden Manuscript is proved by the cross of
Palenque, where we find in signs A B 16 precisely the date I Ahau 18 Zotz, a Tonalamatl before 18
Kayab, in D 1 C 2 exactly the difference 2200 and in D 3 C 4 the date IV Ahau 8 Cumhu. This is in favor
of the theory that our Manuscript did not originate far from Palenque.

Now, the question finally arises as to what may, strictly speaking, be considered the significance of
these numbers, dates and differences.

In the first place, I would recall the fact that the dates of the monuments of Copan and Quirigua, which
doubtless refer to present time, are in the neighborhood of 1,400,000. The high numbers of our
Manuscript, so far as they are in question here, form first a group, which extends from about 1,200,000
to 1,280,000, and then there is a blank, and next a large group extending from about 1,350,000 to
1,480,000, then another blank and lastly a group extending from about 1,520,000 to 1,580,000. If we
assume that our Manuscript belonged to about the same date as these inscriptions, then the three
numbers discussed here would extend over a past period lying about 160-170 years back, when a new
period of importance had begun probably dating from the immigration of the Aztecs into Mexico, which
they placed in the first half of the 14th century (see "Weltall," Vol. 5. pp. 374-377). Now, however, the
number 1,366,560 contains the statement that 3744 years ago (each year having 365 days) an event
must have occurred, which can hardly be anything other (according to the belief of the Mayas) than the
creation of mankind. Hence all the historical dates of the Mayas were computed from this starting-point.
But how did this event come to have the date IV Ahau 8 Cumhu?

In my opinion this date is to be regarded only as the result of the far more important date I Ahau 18
Kayab, lying 2200 days earlier. Day 17, Ahau, belongs, without doubt, to the chief of the gods, and as
the first week day it must have been especially sacred. The prophecies of the Tonalamatl preferably
begin with the Ahau and with the I. The series on the page under discussion, constructed with the
difference 2920 as a basis, begins with I Ahau, and the three series on pages 46-50 also have the same
day as the zero point of departure. I Ahau is therefore the starting-point of the astronomical
computations as IV Ahau is of the historical.

Now, however, all the periods of 260 days end each time with I Ahau. Why is precisely this day chosen
here, which is the 18th day of the month Kayab, therefore in the year 3 Kan, and lying 2200 days earlier
than the historical date?

Day 18 Kayab is our June 18th. In my treatise "Schildkröte und Schnecke in der Mayaliteratur" (1892), I
have sought to prove that the tortoise served as symbol of the summer solstice, that the sign of Kayab
was the head of a tortoise, and that probably the 18th of June was regarded as the longest day. The
middle one of the three series on pages 46-50 begins with exactly this date, I Ahau 18 Kayab.

But whence come the 2200 days? I will offer a suggestion which may serve until a better theory is
propounded. Let us assume that each of the five principal planets had in succession regulated its time
of revolution by this astronomical starting-point, thus:—sun 365, moon 356, Mercury 115, Venus 584,
and Mars 780 days, these numbers added together give exactly 2200. It will scarcely excite surprise
that I should set down the lunar year at 356 days (and not at the usual 354 days) for there are 12 ×
29⅔ lunations in a year and we thought we had already found this period on this page, while
discussing the number 1,364,360; also on pages 51-58, in addition to the half lunar year of 177 days,
we shall find one of 178 days. Were the planets therefore created 2200 days before the appearance of
mankind? Jupiter and Saturn, of course, with their 397 and 380 days are probably not considered here,
because their periods of revolution so nearly correspond to that of the sun, and on pages 51-60 they
are also treated as of secondary importance.
I confess I am quite unable to discover what may have happened 11,960 days before the creation of
the stars—possibly the birth of one of the principal deities. Perhaps one of my fellow-students may
succeed in finding an answer in one of the creation myths.

We come now to the 40 hieroglyphs on the left half of the page. These are intended simply to
familiarize the reader with those signs which are of importance in the calendrical-astronomical portions
of the Manuscript. Since no phonetic system of writing existed, we cannot, of course, expect that the
scribe should have explained these signs.

Signs 1-4, which are mostly destroyed, can hardly denote anything other than the four quarters of the
globe, at least we can still recognize in 4 the sign for the east, which has also the fourth place in pages
46-50. They stand thus together five times in the middle of the left side of pages 46-50, which pertain
to this subject. 5 to 9 are the sign for Venus repeated 5 times, probably denoting the four parts of its
revolution as on pages 46-50 and also the revolution as a whole. In connection with this first
appearance of the Venus sign, I would mention that the same hieroglyph also occurs in the Tro-Cort.,
e.g., Cort. 25c, though this Manuscript contains little else that is astronomical, yet it also has the
rectangular heavenly shields.

10. This is a well-known form of the Moan sign. In the Globus, Volume LXV, 1894, I sought to make it
appear probable that the Moan also denoted the Pleiades, with whose disappearance and reappearance
the beginning of the years seems to be connected. Likewise on page 50, where the 2920-period ends,
the Venus and Moan signs appear at the top on the right-hand side.

11 and 12 are the same sign, being that of the 13th Uinal (Mac), with which 260 days of the year end,
and hence this sign is also used as the sign of the Tonalamatl. The repetition seems to show, that not
until the 73 Tonalamatls of the period of 18,980 days are doubled—thus obtaining the number 37,960 of
such importance here—are the sun and Venus periods brought into unison (with the whole system).

13. The Kin sign (sun, day) with the superfix, which in all probability expresses conjunction, union, and
which, in my opinion, we also see on page 51, combined with Kin and Imix, as the sign for 18,980 days,
is used here after the two Tonalamatls to denote the doubling of this period.

14-18. If the preceding signs led us to the Venus-solar period and to the continuation of this subject on
pages 46-50, these five hieroglyphs bring us to the Mercury-lunar period and later, on pages 51-58,
which are devoted to the same period, we shall find a parallel especially on the last page. First comes
14, which, as has been acknowledged, is the sign for 20 × 360 = 7200 days. 15, a hand holding a
rectangle divided by a cross into four parts, is, I believe, the sign for the period of 20 days augmented
to 21 by the 1 in front of it. The much more distinct form of sign 16 on the middle of page 58 and also
at the top of page 53, should be compared with the sign as given here. The top part is the familiar Ben-
Ik sign denoting the 10th and 19th days, and the bottom is the sign of the 14th division of 20 days,
which make up the year. Now, however, the 10th day, when it becomes the 19th of the next 20 days, is
distant from the first 29 days. The prefix consists of two parts:—First two small circles joined by a
zigzag line, which I think denotes the division of a day into halves; the sign would then equal 29½
days, i.e., very nearly the true lunar month. Second, of two vertical lines, which might denote a
doubling. The whole would then be equal to 2 × 29½ = 59. I admit that this interpretation is very
artificial and I should be very glad if a better explanation could be found. On the other hand the 17th
hieroglyph becomes quite clear when it is compared with the parallel passage on page 58; it is 13 × 360
= 4680 days, a third of the remarkable period of 14,040 days.

Thus we have

Hieroglyph 14 = 7200
" 15 = 21
" 16 = 59
" 17 = 4680
 ———
11960,

which is exactly the lunar-Mercury period.

The sign Xul = conclusion, end, is fittingly added in 18 to the end of this period, as also on page 58.
This sign is very common on pages 61 and 62 at the end of the long periods.

From signs 19 and 20 we see that the four parts of the Venus year are also about to be treated of here,
that is, the periods of 236, 90, 250 and 8 days respectively, which are discussed on pages 46-50. For 19
is the sign for Venus, and 20 is a hand with a knife as a superfix, which divides the Venus revolution.
This hand appears 20 times in like manner on the pages mentioned above.

Signs 21-25 represent five gods, who in all probability are N, F, H, the bat-god and A. These are the
same signs which are repeated twice on the left-hand side of pages 46-50, both times at the beginning
and end of the period of 236 days, that is, the period during which Venus is the morning star and which
is under the dominion of the east. The fact that there is a 4 with N has reference to the four forms
which this Uayeyab god assumes. Now we ought to expect a similar treatment of the periods of the
planet, which are under the rule of the south, west and north, but there is no room for this. Instead, we
find in 26, 27 and 28 three different signs plainly belonging together, the first of which is the day Caban,
i.e., the earth; the second may be Muluc denoting rain and water; the third is Chuen (the ape) which
fittingly denotes the north, for Chuen denotes the little bear, as I have proved in my treatise on the day-
signs of the Mayas. The Chuen sign in 28 also has a prefix, which probably refers to the night-god D. I
find exactly the same combination in signs 8 A and 8 B of the inscription on the Cross of Palenque, but I
must leave to others the task of connecting 26 and 27 likewise with the north, which is very evident in
27 (Muluc).

Sign 29 is entirely effaced. Nevertheless, I am positive that it represented the day IV Ahau, the
beginning of Maya chronology, for 30 may still be identified as 8 Cumhu belonging to IV Ahau, and sign
31 is the same sign as 18, i.e., the sign Xul = the end, and denoting here the end of the long period.

The comprehensive hieroglyphs, 29-31, stand here in the wrong place. A more suitable position for
them would be before 19 or just after 35. For they are intended to specify the periods during which
Venus is in the west and south, i.e., the time during which it is the evening star and the period of its
inferior conjunction.

Sign 32 is the black deity, L according to Schellhas, here denoting the west, and 33 is the Venus sign
with the prefix denoting division. In the same way we find these two signs together on page 46 at the
right in the middle series, where presumably the four Venus periods are specified in close succession.
The black deity is also found on page 50 in the middle of the page in the beginning, at the end of a
period of 250 days. On page 24 it has as a prefix the sign Imix with three rows of dots proceeding from
it. Imix, however, among the Mayas and Aztecs (as Cipactli), under some circumstances often, and
under others always, denotes the first of the 20 days. Hence this sign may mean:—here begins the
Venus period of 250 days.

34-35. The sign for the south still remains to be found. Sign 35 is again the Venus hieroglyph. In 34 we
should expect to find one of the five gods of the south, which are found on pages 46-50, e.g., the
Moan, who is represented on page 47 as the regent of this cardinal point. But there is no figure of a
god here, and in place of it we find set down here, as on page 47, middle, right-hand, an actual date as
the beginning of this short southern period of only eight days. It is the date 10 Zip (third month), the
month sign of which does indeed suggest a hieroglyph of the Moan. Now, if we recall that in hieroglyph
21 the god N is designated in exactly the same way by an actual date, viz:—4 Zac (11th month), then
we see that the interval between 4 Zac and 10 Zip of the second year following, is exactly 236 + 90 +
250 = 576 days, and this corresponds exactly to the interval of time from the beginning of the period
when Venus is in the east to the beginning of the period when she is in the south. If we knew in what
years the morning star made its first appearance on February 4th and disappeared as the evening star
on the 3d of September, we should make some progress in the comprehension of this subject, but not
much, since these events fall approximately on the same dates after each period of 8 years.

36-40. The last five of the 40 signs appear in the same order again on pages 46-50, one sign on each
page, in the middle group of the right-hand half of the page at the beginning of the third line, but with
this difference, that on page 24 each sign has the same prefix, which is wanting on pages 46-50, where
a similar hieroglyph always follows. From their position on pages 46-50 it follows that these are
hieroglyphs of five gods, each of whom belongs to a whole Venus year of 584 days. I am not very sure
in regard to these gods. I prefer to call 36 K, 37 F, 38 E and 40 A. Sign 39 with the person crouching, I
am obliged to leave entirely unsettled. We shall find this hieroglyph again, e.g., on pages 47 and 49
right, middle. Let it suffice that in these five signs we have a repetition of the Venus-solar period of
2920 days, with which we will end the discussion of this page. Only F and A have already been met with
among the five gods denoted by hieroglyphs 21-25.

Pages 25—28.

As these four pages, which are the beginning of the back of the first part of the Manuscript, not only
belong together, but also display a parallel arrangement of their separate parts, the corresponding parts
will be considered together as a whole.

There are seven of these parts on each page, viz:—the column of day-signs on the left hand; the top,
middle and bottom pictures, and lastly the top, middle and bottom groups of hieroglyphs; but I will
consider the pictures and hieroglyphs of the same section as belonging together.

1. The Columns of Day-Signs.

On the left-hand side of each page two days are repeated 13 times. They are as follows:—On page 25
Eb and Ben, on page 26 Caban and Ezanab, on page 27 Ik and Akbal, and on page 28 Manik and
Lamat. Cyrus Thomas first made the important discovery that these pages represent the transition from
one year into the next, but held the erroneous opinion that the last two days of each of the four kinds
of years were treated of on each page. While Seler, on the other hand, found that we have here to do
with the last day of one year and the first of the following year, and that, therefore, Ben, Ezanab, Akbal
and Lamat are the beginnings of the years and thus of the 20-day periods. The years, however, were
always named after their second day (i.e., Kan, Muluc, Ix and Cauac years), since the New Year's Day
was considered unlucky and it was the practice of the Mayas to conceal the real starting-point.

These four pages, therefore, extend over 13 × 52 years, that is, over a period of 18,980 days, after
which period all the calendar dates are repeated. A list of all these dates is given in "The Maya and
Tzental Calendars" by William E. Gates (Cleveland, 1900).

The transition from the Muluc to the Ix years is represented on page 25; from the Ix to the Cauac years
on page 26; from the Cauac to the Kan years on page 27, and from the Kan to the Muluc years on page
28. The Ix years are represented first, because the beginning of the historical chronology lies in an Ix
year (IV Ahau; 8 Cumhu). This section treats of ceremonies, especially of the setting up of the idols at
the changing of the year, which I can pass over here since they have already been described by Diego
de Landa and in our own day by Cyrus Thomas in his "Study of the Manuscript Troano," and elsewhere.

2 . T h e To p P i c t u r e s .

The principal representation on all the four pages is a priest, but disguised as an animal with the head
of a beast of prey as a mask (always the same one) and also with a tail. He is pictured with the same
three articles in each of the four representations, viz:—First, in his right hand, the staff of office with the
hand at the top, which, according to Seler, "Mittel-Amerik. Musikinstrum.," p. 112, is the rattle-stick,
second the incense-pouch, i.e., for copal, and third in his left hand a rattle, or, according to Schellhas,
Welcome to our website – the perfect destination for book lovers and
knowledge seekers. We believe that every book holds a new world,
offering opportunities for learning, discovery, and personal growth.
That’s why we are dedicated to bringing you a diverse collection of
books, ranging from classic literature and specialized publications to
self-development guides and children's books.

More than just a book-buying platform, we strive to be a bridge

connecting you with timeless cultural and intellectual values. With an
elegant, user-friendly interface and a smart search system, you can
quickly find the books that best suit your interests. Additionally,
our special promotions and home delivery services help you save time
and fully enjoy the joy of reading.

Join us on a journey of knowledge exploration, passion nurturing, and

personal growth every day!

ebookbell.com