The Origin Of Agent Markers Enrique L Palancar

The Origin Of Agent Markers Enrique L Palancar
The Origin Of Agent Markers Enrique L Palancar
The Origin Of Agent Markers Enrique L Palancar

1. The Origin Of Agent Markers Enrique L Palancar
download
https://ebookbell.com/product/the-origin-of-agent-markers-
enrique-l-palancar-51135418
Explore and download more ebooks at ebookbell.com

2. Here are some recommended products that we believe you will be
interested in. You can click the link to download.
Agents Of Reform Child Labor And The Origins Of The Welfare State
Elisabeth Anderson
https://ebookbell.com/product/agents-of-reform-child-labor-and-the-
origins-of-the-welfare-state-elisabeth-anderson-51956074
Agents Of Reform Child Labor And The Origins Of The Welfare State
Princeton Studies In Global And Comparative Sociology Elisabeth
Anderson
https://ebookbell.com/product/agents-of-reform-child-labor-and-the-
origins-of-the-welfare-state-princeton-studies-in-global-and-
comparative-sociology-elisabeth-anderson-37581202
The Origin Of Chinese Characters An Illustrated History And Word Guide
Kihoon Lee
https://ebookbell.com/product/the-origin-of-chinese-characters-an-
illustrated-history-and-word-guide-kihoon-lee-46462488
The Origin Of Empire Rome From The Republic To Hadrian 264 Bc Ad 138
David Potter
https://ebookbell.com/product/the-origin-of-empire-rome-from-the-
republic-to-hadrian-264-bc-ad-138-david-potter-48359844

3. The Origin Of Oughtness A Case For Metaethical Conativism Fischer
https://ebookbell.com/product/the-origin-of-oughtness-a-case-for-
metaethical-conativism-fischer-49155276
The Origin Of Higher Taxa Palaeobiological Developmental And
Ecological Perspectives T S Kemp
https://ebookbell.com/product/the-origin-of-higher-taxa-
palaeobiological-developmental-and-ecological-perspectives-t-s-
kemp-49432318
The Origin Of The Modern Syrian Kalil Bishara
https://ebookbell.com/product/the-origin-of-the-modern-syrian-kalil-
bishara-49445304
The Origin Of The Doric Entablature Analecta Gorgiana 1st Edition
Leicester Holland
https://ebookbell.com/product/the-origin-of-the-doric-entablature-
analecta-gorgiana-1st-edition-leicester-holland-49450162
The Origin Of The Recessive Accent In Greek Maurice Bloomfield
https://ebookbell.com/product/the-origin-of-the-recessive-accent-in-
greek-maurice-bloomfield-49450274

5. Palancar
The Origin of Agent Markers

6. Studia typologica
EDITED BY THOMAS STOLZ
Beihefte / Supplements
STUF - Sprachtypologie und Universalienforschung
Language Typology and Universals
VOLUME 5

7. Enrique L. Palancar
The Origin of Agent Markers
Akademie Verlag

8. Die Deutsche Bibliothek - CIP-Einheitsaufnahme
Palancar, Enrique L.:
The origin of Agent Markers / Enrique L. Palancar. - Berlin :
Akad. Verl., 2002
(Studia typologica; Vol. 5)
ISBN 3-05-003767-9
ISSN 1617-2957
© Akademie Verlag GmbH, Berlin 2002
Das eingesetzte Papier ist alterungsbeständig nach DIN/ISO 9706.
Alle Rechte, insbesondere die der Übersetzung in andere Sprachen, vorbehalten. Kein Teil des Buches
darf ohne Genehmigung des Verlages in irgendeiner Form - durch Photokopie, Mikroverfilmung oder
irgendein anderes Verfahren — reproduziert oder in eine von Maschinen, insbesondere von Datenver-
arbeitungsmaschinen, verwendbare Sprache übertragen oder übersetzt werden.
All rights reserved (including those of translation into another languages). No part of this book may be
reproduced in any form—by photoprinting, microfilm, or any other means—nor transmitted or translated
into a machine language without written permission from the publishers.
Druck und Bindung: Primus Solvero, Berlin
Printed in the Federal Republic of Germany

9. A la memoria de mi madre,
Clara Vizcaya Martinez,
en honor a su alma, a su sonrisa,
a su lengua de agua,
y a sus raices en la antigua y legendaria Galicia

11. Contents
1 Introduction 1
1.1 Positions to take concerning Agent case-syncretisms 2
1.1.1 The Syntactic Position 3
1.2 The purpose of this book: The semantic position 4
1.3 Case-syncretism as an object of study 4
1.4 Functional extensions 6
1.5 Some fundamental theoretical assumptions 7
1.6 A glance to Agent case-syncretisms in the literature 7
1.7 What you can skip reading: The skeleton of the book 8
2 Defining the Object of Study 10
2.1 The Agent 10
2.1.1 Passive Agents 11
2.1.2 Ergatives 11
2.2 Agent marking NOT treated in the study 13
2.2.1 Other unmarked Agents 14
2.2.2 Active and Pragmatic Agents 15
3 The Samples 16
3.1 The Sample of Agent Markers or Main Sample 16
3.2 The General Sample 20
4 Asyncretic Agent Markers 22
4.1 Further contrasts: One or more markers to express Agent 23
4.2 Languages and markers 24
4.3 Summary 24
5 Syncretic Agent Markers 26
5.1 Categories involved in the syncretisms 26
5.1.1 Cause 27
5.1.2 Causee 29
5.1.3 Dative 30
5.1.4 Intermediary 31
5.1.5 Instrument 32
5.1.6 Possessor 33
5.1.7 Ablative 33
5.1.8 Allative 35
5.1.9 Locative 36
5.1.10 Perlative 37
5.1.11 Categories with a secondary involvement 38
5.2 About the syncretisms 39

12. 5.3 Syncretic Agent markers 40
5.3.1 Involvement of semantic categories 40
6 Functional Extensions and Agent Markers 45
6.1 Functional extension vs. grammaticalization 46
6.2 Agent markers in the literature 47
6.3 Space first! ... but always? 48
6.3.1 Shortcomings of localist accounts 51
6.4 Alternatives to the localists: LEHMANN's proposals 52
6.5 What about other thematic categories: The case of Cause 54
6.6 Other approaches: The semantic alternative 56
6.6.1 The Agent as a prototypical Instrument: WIERZBICKA's proposal 57
6.6.2 Transformations from an Instrument: JANDA's proposal 57
6.6.3 Some brief insights: DOWTY's proposal 61
6.7 Summary: What position to take? 61
7 Causation 63
7.1 Causation archetypes 64
7.1.1 The causation archetype in a morphodynamic model 68
7.2 Role arquetypes: The Agent 69
7.3 TheEnergizer 70
7.4 Conceptualizing Energizers 75
7.5 Construals and grammar: The transitive construction 78
7.5.1 Inverse construal: The passive 83
7.5.2 Other aspects emerging from the construals 86
8 Instruments 88
8.1 The Instrument in Case Grammar 88
8.1.1 Discussion 91
8.2 A morphodynamic characterization of the Instrument 94
8.2.1 The macro-event and the medial-event 99
8.2.2 The Medial-Entity 102
8.2.2.1 Instrumental reading of the Medial-Entity 102
8.2.2.2 Enabling reading of the Medial-Entity 103
8.2.3.3 Causal reading of the Medial-Entity 104
8.3 The dynamic understanding of the Instrument 105
8.4 The Medial-Entity as an Object of a verb 108
8.5 More examples of pregnance 112
9 Cause 115
9.1 Cause as an Energizer 115
9.2 The syncretism of Cause with other categories 118
9.2.1 From Cause to Instrument 121
9.2.2 More syncretisms of Cause and Instrument 124
9.2.3 From Instrument to Cause 126
9.3 Cause as Reason 131
9.4 Summary 138
10 Energizers 140
10.1 The Agent as an Initial-Energizer 141
10.2 The Instrument as a Medial-Energizer 142
10.3 Cause as a Force-Energizer 142
10.4 Energizing categories including reference to an Agent 144

13. 10.4.1 The General-Energizer 144
10.4.2 The Autonomous-Energizer 145
11 Passive Agents 147
11.1 Preliminaries to the extensional paths and clines 148
11.1.1 Methodological criteria 149
11.1.2 Notations used in the paths 150
11.2 The Passive Agent as a Causal/Energizer 151
11.2.1 Syncretisms involving Cause and Passive Agent 151
11.2.2 Syncretisms involving Ablative and Passive Agent 155
11.2.3 Syncretisms involving Locative and Passive Agent 161
11.2.4 Syncretisms involving Instrument and Passive Agent 164
11.3 Syncretisms involving Possessor and Passive Agent 174
11.4 Syncretisms involving Causee and Passive Agent 175
11.4.1 Syncretisms involving Instrument, Causee and Passive Agent 181
11.4.2 Syncretisms involving Intermediary, Causee and Passive Agent 183
11.4.3 Syncretisms involving Dative, Causee and Passive Agent 187
10.5 Syncretisms involving Dative and Passive Agent 191
12 The Origin of Passive Agent Markers 196
12.1 The clines 196
12.2 The causal cline 198
12.3 The Causee cline 199
12.3.1 The "Eastern Cluster" 201
12.4 The Dative cline 201
12.5 The Instrument cline 203
12.6 The Possessive cline 204
12.7 The involvement of spatial categories 205
12.7.1 TheAblative 205
12.7.2 The Perlative 206
12.7.3 The Locative 206
12.7.4 The Allative 207
12.8 Brief summary of the clines 208
13 Ergativity: A conceptual approach 209
13.1 Towards a conceptual characterization of ergativity 211
13.1.1 Antipassive constructions 214
13.2 The continuum of grammatical organization 217
13.3 The origin of ergativity 219
13.4 Ergative syncretisms 224
13.4.1 Ergative and Genitive 224
13.4.2 Ergative, Dative and Locative 225
13.4.3 Ergative and Instrumental 227
13.4.4 Summary of Ergative syncretisms 228
14 Ergatives 229
14.1 The Ergative Agent as a Possessor of the action 229
14.2 The Ergative Agent as a Causal/Energizer 231
14.3 Preliminaries to the reanalysis: The Effector 236
14.4 Syncretisms of Locative and Ergative 241
14.5 Ergatives with other origins 242
14.5.1 The case of Tiriyo 242
14.5.2 The case of Kabardian 243

14. 15 The Origin of Ergative Markers 244
15.1 The extensional paths of Ergative 244
15.2 The clines 245
15.3 From the Causal Stage to the Actional Stage 247
15.3.1 The dynamics leading to the Actional Stage 249
15.3.2 The case of Basque and Trumai 253
15.3.2.1 Basque 253
15.3.2.2 Trumai 255
15.3.3 A brief summary of the dynamics 256
15.3.4 Markers and Stages 256
15.4 The involvement of spatial categories 258
16 Revisiting Agent Markers 262
16.1 Contrasting Agent markers 262
16.1.1 Ratio of syncretism of the markers 262
16.1.2 Available markers per language 262
16.1.3 Morphemic profile of the markers 263
16.1.4 Categories involved in the syncretisms 263
16.1.4.1 Irrelevance of categories 263
16.1.4.2 Absence of categories 264
16.1.4.3 Marginal presence of categories 264
16.1.4.4 Frequent presence of categories 264
16.2 Interpreting the data 265
16.2.1 The Energizer 265
16.2.2 The Direct and the Inverse construals 265
16.2.3 Types of Energizers 265
16.2.4 The construals and grammatical structure 266
16.2.5 Ergatives and Passive Agents 266
16.2.6 The choice of a spatial category 267
16.3 Overall conclusion of the book 268
References 269
Appendixes 294
Appendix 1: Languages in Main Sample 294
Appendix 2: Sample of Passive Agent Markers 297
Appendix 3: Sample of Ergative Markers 299
Appendix 4: General Sample 300
Appendix 5: Syncretic Patterns in Agent Markers 305
Appendix 6: Extensional Paths for syncretic Passive Agent markers 307
Appendix 7: Paths in Indo-European and the World's Languages Sample 308
Appendix 8: Extensional Paths for syncretic Ergative markers 310

15. Acknowledgements
My heart and mind rejoices with the loving memory of many dear people who
helped me in one way or another in the creation of this book. To all I offer my most
sincere thanks. So many people come to mind that I am afraid I won't be able to
mention all of them here.
In the intellectual circle from Madrid, where I wrote the thesis that served as a
preliminary version for the present book, I want to thank my colleague and mentor
Enrique Bernärdez, undoubtedly the best language thinker in the Peninsula, for all
the conversations over a cup of coffee. I am also indebted to Julia Mendoza, the
first to introduce me to the aspirated labials of proto-Indo-European in my first
years in college, and later the intellectual who gave me all her support in difficult
times. To Maika Guarddon and Maria Cristobal for all our clandestine talks, and to
Marisa Blanco, with whom I started the journey to cognitivism long ago.
In Berkeley, I am indebted for life to the teachings of Eve Sweetser, and to Dan
Slobin for all those brilliant verb-framed ideas he scatters about the place while he
moves, but especially for his ability to listen. Here I also convey my gratitude to
my friend Maaike Belien for sharing this never-ending path to knowledge. Dan and
Maaike serve as bridges to the Netherlands. My thanks to the institution of the Max
Planck Institute for Psycholinguistics at Nijmegen, this being a metonymy for the
many brilliant people who make it possible, especially to Melissa Bowerman, one
of the persons I most admire and respect in this intellectual world of ours, always a
sublime inspiration for many of us; also to sweet and profound Penny Brown; to
genius Jürgen Bohnemeyer; to kind and bright Anna Margetts; to shy and thorough
Birgit Hellwig; to grand Eva Schultze-Berndt; to quick Anne Senghas; to unique
Sergio Meira for all our nocturnal discussions; and to chingon Roberto Zavala with
whom I so much enjoy talking about amerindianisms. Out of the Netherlands, but
just across the border, my honest and serious thanks to Thomas Stolz, for his
generosity and typological wisdom, for our common interests in the field, and for
kindly offering me the chance to submit the present book for publication.
In Norway, I want to convey my most profound gratitude to the Oslo
intellectual circle of Rolf Theil Endresen, Hanne Gram Simonsen, Even Hovd-
haugen for treating me so well while I was one of them in the enchanted far North,
and especially to my friend Jan Erik Rekdal, the poet, the intellectual, the human
being for teaching me many other things about life through his beautiful language,
and to Karin Sveen, the poetess, for the bounty of her speech and heart.

16. Back to America, in Mexico, my present home, I am indebted to Ricardo
Maldonado for many things, but especially here for inviting me to this ancient land
of his, and for enabling me to access the treasure of Amerindian languages. To all
my colleagues and friends at the Universidad Autonoma de Queretaro, especially to
Alejandra Auza, Margaret Lubbers and Martha R. Isias and to Phyllis Herring who
kindly proofread the manuscript for me. I also want to express my gratitude to the
Facultad de Lenguas y Letras at the Universidad Autonoma de Queretaro, and
especially to the current dean, Silvia Yreri Mendoza for all the support I received
while writing this book.
My deep thanks to my warden angel on earth, Mariam Collado, probably the
best graphotherapist in the world nowadays, who saved me from the well of
loneliness and drove me to spiritual health. To my best friend ever Virginia
Gurucharri for her grand heart, and to Alex Diez for his understanding. To Maggie
Ford for her Norwegian voice, her tenderness and generosity, and to Pete Levinthal
for his presence and inspiration.
To my parents, my living dear father Honorio Palancar and my deceased dear
mother Clara Vizcaya. They did everything right. They offered everything. I pray
to God to do just as well with my children-to-be as they did with me. And last, but
not least, to Arantza Pascual, my wife, my companion; she is the only living soul
on earth who knows how much I invested in this book, and she was there all along.
Gracias Aran.
And why not?, To God-Goddess, who made all real and unreal possible for us,
for the inspiration of the Muses, and for giving many of us work after Babel.
Needless to say, I am the only one to blame for all errors and shortcomings.
At Queretaro, Mexico, at the time when
the jacarandas blossom, in the year two
thousand and two.

17. Abbreviations
1,2,3 Persons LOC Locative
& My own glosses Μ Masculine
ABL Ablative MPROP Modal Proprietive
ABS Absolutive Ν Neuter
ACC Accusative NHON Non-Honorific
AG Agent NML Nominalizer
APP. Applicative NOM Nominative
APPR Appropriative NPG Negative Perfect Gerund
ART Article NPST Non-Past
ASP Aspect marker NSGA Non-Singular Actor
CL Classifier OBJ Object
COM Comitative OBL Oblique
COMP Completive aspect Ρ Participle
COMPL Complementizer P. Preposition
CONT Continuous aspect P.P Past Participle
CPf Class Prefix PASS Passive
DAT Dative PERF Perfect
DC Declarative Sentence Type PERS Person
DE Direct Evidential Pf. Prefix
DEF Definite PL Plural
DEM Demonstrative PP. Postposition
DESIDERA Desiderative PRES Present
DIR Directional PROP Proprietive
DPR Deprehensative PRT Particle
DU/D Dual PSP Polite Speech Particle
EMPH Emphatic PST Past
ENE Energizer R Reflexive
ERG Ergative REC Recipient
EXT Extensive REL Relative
F Feminine RN Relational Noun
FUT Future RS Relational Suffix
HAB Habitual Sf. Suffix
HON Honorific SG Singular
I. Inflection SIM-CV Simultaneous Co-verb
IMP Imperfect SP Specific
INANA Inanimate Anaphoric SUBJ Subject
INC Inclusive T/A Tense/Aspect
INCH Inchoative TEMP Temporal
INCOMP Incompletive TH Thematized participant
IND Indicative ThV Thematic Vowel
INF Infinitive TNS Tense
INPROX Inanimate Proximate TOP Topic
INS Instrumental TR Transitive
INSTR Instrument U Undergoer
INTR Intransitive VAL Validator
L (Phonotactic/Grammatical) Link

18. Linguistic Families
AA Austro-Asiatic
AF Afro-Asiatic
AL Altaic
AM Amerind
AR Arawakan
AT Australian
AU Austronesian
C Creole
CA Caucasian
CD Caddoan
CH Chibchan
CK Chukotko-Kamchatkan
CR Carib
EA Eskimo-Aleutian
ED Elamo-Dravidian
HK Hokan
IE Indo-European
IP Indo-Pacific
IR Iroquian
JV Jivaran
Κ Khoisan
LI Language Isolate
MJ Macro-Je
MY Mayan
MZ Mixe-Zoquean
ND Na-Den6
NK Niger-Kordofanian
NS Nilo-Saharan
OM Oto-Manguean
PN Panoan
PT Penutian
PY Peba-Yaguan
QU Quechuan
SL Salish
SN Siouan
ST Sino-Tibetan
TG Tupi-Guarani
UA Uto-Aztecan
UY Uralo-Yukaghir
WK Wakashan
YN Yanomam

19. 1. Introduction
Traditionally, case-syncretism applies when two or more distinct morphosyntactic
categories are given identical morphological realization1. Following TRASK, the
phenomenon could be exemplified as follows (both morpheme boundaries in the
examples and boldface are mine):
"In Latin, for example, genitive and dative case forms of nouns are usually
distinct (nominative amic-us "friend', genitive amic-i, dative amic-o), but in
the α-stem nouns they exhibit syncretism (nominative puell-a 'girl', genitive
/dativepuell-ae)" (1993:272).
Case-syncretism may arise from the morphological reduction of a given system
(HARTMANN & STORK 1972:229). This could be illustrated in Old English
(OE) with the merging of the former Nominative-Accusative distinction in α-stem
nouns in Proto-Germanic, a pattern preserved in Old Norse (ON), e.g. OE dceg
"day" (NOM/ACC) vs. ON dag-r (NOM) and dag (ACC).
In this book, contrary to the traditional conception, I will make a special use of
case-syncretism. I take case-syncretism to be the phenomenon that applies when
two or more semantic roles are given the same linguistic realization. In other
words, case-syncretism applies when a given grammatical marker conveys two or
more roles that are treated as independent semantic categories. More specifically, I
concentrate on the case-syncretisms displayed by a variety of Agent markers. In
order to gain a better understanding of the complexity behind this phenomenon, let
us start the discussion with some examples of syncretic encodings. For this pur-
pose, consider examples (1) and (2) from Russian, (KILBY 1986: 324, the glosses
have been slightly modified).
As it will be common from now on, underlining is used for the morphemic
material expressing a given participant, while boldface type is employed to
highlight any grammatical material relevant for the discussion.
1 The term "syncretism" comes from Medieval Greek, but has an uncertain etymology. The original
Greek word is συγκρετισμός [sirjkritis'mos] (in modern reading), from the verb συγκρετιζειν
[siykri'tizin]. The verb referred to the alliance of two parties when fighting against a third. An etymo-
logy dating from the XVIIth century posits that the word referred to a type of alliance common in
Crete, the Greek island (OED, vol. Su-Sz. p.378). The word has as root the name of the Greek island
Κρήτη /'kriti/ and shows the associative prefix συν-/sin/.

20. 2 1. Introduction
(1) on byl prinjat ministr-om
he was received-M minister- INS:M
'He was received by the minister'
(2) ona udarila ego palk-oj
she hit-F him:ACC stick-INS:F
'She hit him with a stick'
In example (1), the head of the NP expressing the Agent in the passive construction
is inflected with the so-called "Instrumental" case. The name "Instrumental" is mo-
tivated here because this inflection typically encodes the semantic role of the
Instrument, as in (2). In this sense, the Instrumental inflection in Russian shows
case-syncretism, given that it serves to encode two distinct semantic roles: an In-
strument and an Agent. The same phenomenon is frequent in other Indo-European
languages (cf. LURAGHI 1987). We witness the pattern in Sanskrit, as illustrated
in examples (3) and (4), (COULSON 1976:58):
(3) jitah Räksasa-h Cänakv-ena
beaten-NOM Räksasa-NOM Cänakya-INS
'Räksasa was beaten by Cänakya'
(4) jal-ena asvän sinc-ati
water-INS horse-ACC:PL sprinkle-3SG:PRES
'He sprinkles the horses with water'
With this in mind, we can further ponder about the motivation for syncretisms like
these. The choice of the "Instrumental" case in these related languages might be (at
times is often) understood as an encoding strategy inherited from the proto-
language (LEHMANN 1974:190 ff.). This type of explanation is legitimate only in
a rather restricted sense because there are many languages of the family that pre-
sent different patterns. More importantly, the same syncretism is commonly found
in other unrelated languages. The pattern is for example seen in Northern Tepehuan
(UA:AM) as shown in (5) and (6), (BASCON 1996):
(5) ikiityi Piili gi-növi imo kuxiiru-i-kidi
cut Felipe 3-hand a knife-ABS-INS
'Felipe cut his hand with a knife' (291)
(6) gi-uii go-baabmdyi imo gogooxi-kidi
PASS-eat the-meat a dog-INS
'The meat was eaten by a dog' (295)
1.1 Positions to take concerning Agent case-syncretisms
We may understand this phenomenon in at least four different ways. The different
possibilities are presented as follows:

21. 1.1 Positions to take concerning Agent case-syncretisms 3
- Arbitrariness of the system: The syncretisms are seen as mere coincidences
of the system involved.
- Genetic origin: The syncretism is accounted for as being the outcome of a
common proto-language.
- Syntactic position·. The phenomenon is observed under a syntactic
perspective. The selected marking encoding the Agent is interpreted as an en-
coder of an oblique constituent, not of a semantic role. Cross-linguistic com-
monalties are accounted for in these terms.
- Semantic position: The syncretic pattern is motivated by semantic links op-
erating across the categories involved in the syncretism. The semantic links
reflect the application of cognitive processes common to all humans.
The first position has no explanatory power; relying too much on randomness
greatly weakens the understanding of the morphogenesis of human languages. The
second position, while admitting the existence of the phenomenon, may be rejected
by using counter-evidence from other related languages that display different pat-
terns. The third and the fourth position deserve special attention, and I will
examine them in more detail in the next sections.
1.1.1 The syntactic position
A syntactic position is a rather strong stand. It renders satisfactory results if clausal
organization is conceived as the mere result of a variety of abstract syntactic
operations, regardless of their modus operandi. This is typically the position taken
by practically all formal models of syntax: for example Government and Binding
Theory in CHOMSKY (1981), Minimalism (CHOMSKY 1995), Lexical Func-
tional Grammar (BRESNAN 1982), Head-driven Phrase Structure Grammar
(POLLARD & SAG 1993). However, such a position is also accepted in more
functionally oriented models, such as Role and Reference Grammar (FOLEY &
VAN VALIN 1984; VAN VALIN 1993; VAN VALIN & LAPOLLA 1997), Func-
tional Grammar (DIK 1997), or even Construction Grammar (FILLMORE & KAY
1993, 1994; GOLDBERG 1995).
Take for example the case of the encoding of an Agent in the passive. All the
aforementioned models, however different their theoretical assumptions might be,
share a common understanding of the encoding of the NP expressing the Agent in
the passive construction of a given language. The NP expressing this participant is
understood to be an oblique phrase that does not bear constituent status to the
verbal core. What is important here is that the Agent phrase is plainly understood
to be a mere oblique phrase. In light of this, the motivation of the use of Ablative,
Locative, Instrumental, or of whatever other markers to encode the Agent NP is
straightforward given the assumption that such grammatical markers are not under-
stood to be encoders of spatial schemas or of Instruments, but to be encoders of
oblique phrases. The semantics, hence the case-syncretism, of such markers is
secondary. No further pondering is needed for this phenomenon; the demotion of

22. 4 1. Introduction
the Agent phrase to a chömeur position is the only relevant issue. Whether the de-
motion is motivated by pragmatic forces or by organizational principles in the lan-
guage is another matter that each model will solve attending to their own the-
oretical premises. Nonetheless, an approach to Agent case-syncretisms based strict-
ly on a syntactic position has serious shortcomings.
1.2 The Purpose of this book: The semantic position
The NP expressing the Agent in the passive has, no doubt, an oblique status in the
clause; the morphology it receives is marked with respect to other less marked con-
stituents. I will not argue against the marginal syntactic status of such a phrase.
However, I will propose that the choice of one or another marker to express this
oblique participant is not random in the sense that "whatever" marker will do the
job. I will posit that there is some conceptual and semantic motivation for the
choice of certain markers over others to do this job.
There is some evidence for this semantic position that weakens a position based
exclusively on syntactic principles. If the oblique character of the Agent phrase
were the only important factor at work in the organization of the clause, we would
expect that all makers encoding obliques would be equally optimal encoders of
Passive Agents. However, this simply does not happen. There are categories that
are never found as encoders of Agents: e.g. Ilatives or Adessives, or even Comi-
tatives alone (cf. STOLZ 1996). Additionally, as we will see in Chapter 5, the
involvement of certain categories is clearly favored over others: e.g. Ablatives over
Perlatives. Such data constitute evidence for the existence of certain preferences, of
certain cross-linguistic patterns. Furthermore, this preference needs be motivated
by common principles operating across languages. In other words, the motivation
for the encoding of a Passive Agent -and I extend it to Ergatives- needs not be a
pure syntactic phenomenon.
This book is intended to be a contribution to fill this theoretical gap. I provide an
analysis of Agent markers that is oriented toward gaining an overall understanding
of the principles governing the selection of certain encoders over others to express
the Agent in certain constructions. More importantly, the analysis points toward
gaining a better understanding of the way such encoders eventually become Agent
markers.
1.3 Case-syncretism as an object of study
Obviously, the present study originates from the belief that case-syncretism is
indeed an interesting enough phenomenon to become the object of a linguistic
analysis. The reason being is that the case-syncretism of a given marker is taken to

23. 1.3 Case-syncretism as an object of study 5
reveal the existence of a complex interweaving of relations across the categories
involved in the syncretism. In view of this, I depart from the theoretical assumption
that there is some polysemic link across such categories. A polysemic approach to
language phenomena is nowadays well known in linguistic theory (among others,
HAIMAN (1978), LINDNER (1981), BRUGMAN (1983), HERSKOVITS (1986),
LAKOFF (1987), CROFT et al. (1988), HEINE 1990, NIEPOKUJ 1994,
PALANCAR (1997a), COPELAND 2000, IKEGAMI 2000, etc.). Particularly
relevant to this study has been its application to the study of case-syncretism as
found in the important work by CROFT (1991), which is entirely foundational for
the present purpose, and also by SCHULTZE-BERNDT (1993), LURAGHI (1995,
2001), and STOLZ (1996, 2001).
A polysemic position assumes that categories in human cognition display
prototype organization. LAKOFF (1987) emphasized the relevance of
understanding the radial array of categories around a prototype. In cognitive
approaches, in contrast to the common practice in formal semantics where
categories are conceived of as well defined conceptual entities, categories are
understood to have fuzzy boundaries; that is, to present an unstable configuration.
This means that categories are seen as dynamic conceptual entities, which allow
room for linguistic change on the domain of their linguistic representation. The
different nodes that constitute a given category may undergo extensions to the
space of other prototypes, which serve as attractors.
The phenomenon I will be studying in this book concerns the category of
Agent. Chapter 7 will be entirely devoted to understanding this category within the
larger frame of causation. As for Agent case-syncretisms, there is some valuable
literature on the topic that will be revisited in Chapter 6. However, as a way to
show the motivation behind the present study, ANDERSON draws attention to the
gap existing in linguistic theory concerning the relation between Instruments and
Agents:
"In many ergative systems, the ergative marker and the instrumental are
identified; in other systems, such as Estonian, instrumental and comitative are
identical (cf. the English preposition with); elsewhere, comitatives seem to be
related (at least historically) to co-ordinators. Much in this area, too, remains
controversial" (1994a: 453).
As I pointed out, an Instrumental encoding is traditionally seen as a case of an
oblique or circumstantial encoding, ANDERSON draws again attention to our poor
understanding of the role circumstantials play in the predication:
"Among the many uncertainties and contentious issues surrounding notions of
case, some of which are surveyed here, perhaps the least explored is the
character and status of circumstantials. [...] But here in particular much
research remains to be done (in any framework)" (1994b: 459).
I intend the ideas in this book to serve as a contribution to the understanding of the
way thematic roles are conceived of in human language and mind. Case-syncretism
is manipulated here as an optimal theoretical tool to understand the conceptual

24. 6 1. Introduction
relation between the Agent and other semantic categories such as Instruments and
Causes so that a given grammatical marker is able to express these categories
together.
1.4 Functional extensions
When a given marker becomes able to express reference to an Agent, the marker is
commonly understood to undergo a functional extension. Such an extension is
taken to happen from another semantic value present in the marker. This means
that the analyst has to decide which value, among the ones conveyed by the
syncretism of the marker, serves as platform for the agentive extension. Like other
authors, in my own account for the phenomenon I will take a number of theoretical
positions concerning which categories are relevant for this agentive extension.
However, I have observed that it is rather common in previous accounts to
select the category that is most commonly associated with the linguistic name the
marker receives as the extensional platform. In other words, if the marker is called
an "Instrumental", authors -more often than not- take the "Instrument" as being the
only possible way to account for the agentive extension. This constitutes a serious
problem for any analysis given that such a terminological myopia impedes seeing
unbiasedly the phenomenon. This in turn might hinder seeing other semantic values
present in the syncretism that could be playing a more direct role in the extension.
As an illustration of what I am trying to convey, consider the semantic value of
Cause present in the Instrumental marker used in Russian in (7) and in Northern
Tepehuan in (8):
(7) Ivan-a pridavil-o sosn-oj
Ivan-ACC crushed-SG:N pine:tree-INS:F
'Ivan was crushed, inexplicably, by a pine tree' (WIERZBICKA 1980:48)
(8) siilikidi tigntyo gin-kii imo miixityu,
truly need I-house a cat
alii oidyääga popoosiki kääx-kidi
much there's mice that-INS
'My house needs a cat a lot because there are many mice'
(BASCON 1996: 299)
The Instrumental marker in these languages serves as encoder of the Passive
Agent, as illustrated in examples (1) and (6) above. I would argue in Chapter 11
that this Causal value reflects the application of a more abstract category in terms
of which the agentive sense is construed. If the analyst does not see beyond the
conventional name given to the grammatical marker, thematic nuances such as the
one expressed in examples (7) and (8) may pass overlooked. This of course has its
dark side, as everything in life.

25. 1.5 Some fundamental theoretical assumptions 1
In the gathering of the sample of Agent markers, I at times faced the important
question whether the source of my data was presenting all possible extensions of
the marker under study. The sample I used for the study suffers at times from im-
precisions coming from some of the material employed as my source, but also from
my own possible mistakes at gathering the data.
1.5 Some fundamental theoretical assumptions
Agent case-syncretisms are taken to be the reflection of a chain of functional
extensions happening across various semantic categories, which eventually led to
agency. The purpose of the study is to account for the various paths leading to this
agentive extension. In this way, I have taken a number of theoretical assumptions
that are common to cognitive approaches to language. Consider, for example, the
following positions in GOLDBERG (1995:3/4):
1. Grammar does not have any transformational component. Semantics is as-
sociated directly to surface structure.
2. Meaning is based on the conceptual construal the speaker has of situations,
and not on computations of truth conditioned possibilities.
According to (1), a given grammatical realization reflects semantic aspects or
spaces of conceptual meaning. In this way, case-syncretism is taken to reveal a se-
mantic reality. Following (2), the general purpose is to gain an understanding of the
close relation between language and cognition, as the structure of the former helps
to understand the structure of the latter. In a cognitive approach to language, the
role of the speaker as conceptualizer is fundamental. The speaker construes reality
and constructs meanings. The idea of including a conceptualizer in the analysis
goes back to a semiotic tradition stemming in the work of C.S. PIERCE (cf.
SHAPIRO 1983). Under this light, the syncretic phenomenon operating on the
morphological level will be taken to be the direct reflex of a number of conceptual
similarities found by speakers across the categories involved in the syncretism2.
1.6 A glance to Agent case-syncretisms in the literature
The purpose of this study is rooted on previous interest on Agent case-syncretism.
The phenomenon has been often treated in passing, but is commonly observed
when Ergative markers are at stake. Nevertheless, Agent case-syncretism is merely
mentioned in most cases as a descriptive fact, which means that the emphasis is not
2 However, one has to be aware that syncretism may also emerge in a semantically unmotivated
fashion from phonological erosion or loaning.

26. 8 1. Introduction
laid on how the Agent extension takes place, but on the fact that it simply takes
place in the language. DIXON (1994:57-58) is a clear example of such a descrip-
tive position, i.e. he mentions that Ergatives often syncretize with Instrumentals,
Locatives, Genitives, or generalized oblique markers.
Being such a descriptive stand a common procedure when talking about the
syncretism of Ergative markers, authors become less interested in syncretisms
when Passive Agent markers are at stake. Instead, the general interest seems to lie
on commenting how frequent the Agent phrase is. Alternatively, authors may show
interest on the special semantics the marker conveys. Take the case of DIXON for
the sake of citing the same author, who ponders about English by:
"An underlying transitive subject may almost always be included in a passive
construction, marked by the preposition by. This can indicate the role that was
primarily responsible for the activity that has resulted in the state described by
the passive verb. In fact, a by phrase is only included in a smallish minority of
actual passive clauses, in most styles of English" (1991:55).
This could be taken as a paragon example for the general attitude toward this
phenomenon. For Passive Agents, it is common to find a descriptive stand similar
to the one formulated as "for the expression of the Agent in the passive of language
X, the marker Y is used". In any case, there are various fundamental contributions
aimed to gaining a better understanding of the reason why a given morphology ap-
plies to encode Agents. I could mention here some nice works, to name just a few,
ANDERSON (1977), TRASK (1979), LEHMANN (1982, 1988), and more
recently CROFT (1991) or LANGACKER (1991a/b). I will review in more detail
some of this literature in both Chapter 6 and 13.
1.7 What you can skip reading: The skeleton of the book
The present book is divided into three parts:
- Part I, comprising Chapter 2 to Chapter 5, introduces the data on which I
base my analysis. I also include here some preliminary tendencies shown by
these data, especially Chapter 4 and 5. Chapter 5 includes working definitions
of the semantic categories present in the syncretism. The reading of this part is
not necessary; it can be used as reference.
- In Part II, I present the theoretical premises on which I based my analysis of
the data presented in Part I. Part II comprises Chapter 6 to Chapter 10, but
Chapter 13 should also pertain here given its theoretical content (mainly
ergativity). Chapter 6 is devoted to offer a brief review of the literature deal-
ing with Agent extensions. In Chapter 7, I introduce some novel ideas and
representations about the phenomenon of causation, and I introduce here the
notion of an "Energizer" as being a highly abstract energetic category, which

27. 1.7 What you can skip reading: The skeleton of the book 9
serves to understand the category of an Agent. Chapter 8 and 9 are devoted to
give a characterization of roles such as Instrument and Cause in the light of
the Energizer presented in Chapter 7. Chapter 10 is a revision of the different
types of Energizers one may expect across languages.
- Part III is devoted to the analysis of the data in the samples. Chapter 11 and
12 deal with Passive Agent markers, while Chapters 13, 14 and 15 with
markers for Ergative Agents. Chapter 11 and 14 present the different paths
leading to the agentive extension. The reading of these chapters is not entirely
necessary to understand the proposal of this book; the same applies to the first
part of Chapter 5. All these chapters can be used as sheer reference, especially
Chapter 11, which is too long for my taste, event though its length is justified.
In Chapters 12 and 15,1 present an analysis of the paths presented in Chapter
11 and 14. Here is where typological trends are observed, and I recommend
their reading for the interesting results.
The book is further completed with Chapter 16, which presents a summary of the
claims made throughout the study together with the tendencies found in the anal-
ysis of the data. At the end of the book lie the Appendixes comprising the samples
used in the analysis.
I honestly hope you enjoy reading this book, at least part of it, I thought of you a
great deal while writing it up. I also present my apologies for all errors and im-
precisions.

28. 2. Defining the Object of Study
In this chapter, I will present a brief description of the object of study of this book.
This characterization is intended to create a uniform metalanguage for the analysis
of Agent markers that follow in the next chapters.
As I mentioned in the introduction, I have examined here the syncretic
behaviour of markers for Passive and Ergative Agents. These markers are treated in
linguistic theory as renditions of the same semantic role, that of the Agent,
regardless of the different grammatical context in which this category is instan-
tiated. I will not deviate from the standard view at this point, and I have employed
the common conception ascribed to the Agent category. However, I will later mod-
ify this conception in light of the data gathered in my samples. I have taken the
theoretical assumption common in cognitive and other functionalist approaches to
language, that differences in form correlate with differences at the semantic and
conceptual level. Accordingly, divergences concerning case-syncretism in markers
for both Passive Agents and Ergatives are taken to point to subtle differences con-
cerning the way the Agent category is conceived in these different frames. Char-
acterising such differences constitutes one of the major goals of the study.
The contents of this chapter are surely obvious to the language pundit, and their
reading is not entirely necessary. Nevertheless, I consider this chapter relevant as a
way to establish the concepts I have employed in the manipulation of the data. I
will first present the standard working definition of an Agent, and then present my
definition of markers for Passive and Ergative Agents. Similarly, in § 2.2, I will
briefly account for other types of Agent markings that are not considered in the an-
alysis.
2.1 The Agent
The category of Agent I have employed in the analysis corresponds, in general
terms, to the one found in standard linguistic theory stemming from the proposals
in FILLMORE (1968). In the most common case, an Agent is the role of the par-
ticipant instigating and bearing control upon a given action. This action leads to an
event where another participant undergoes a relevant change of state. The other
participant typically plays the role of a Patient. In the prototypical case, the Agent

29. 2.1.1 Passive Agents 11
participant is construed as a human being, and additionally, it may be thought to
have acted intentionally and volitionally. In accord with this definition, I have
studied the markers that render this participant in the following grammatical con-
texts.
2.1.1 Passive Agents
The Passive Agent (PASS-AG) is the type of grammatical realization the Agent
receives when this participant is specified within the passive construction of a
given language. Being that passives are typically intransitive constructions, the
Agent is in most cases encoded as an oblique constituent, which bears no direct
core-argument status to the verb. Examples (1) and (2) illustrate the category in
Jacaltec (MY:AM) and Malayalam (ED), respectively:
(1) x-0-tz 'ah-ot] te' nah χ-u naj
ASP-3SG:ERG-paint-PASS CL house 3:ERG-by CL:he
'The house was painted by him' (CRAIG 1977:77)
(2) kuftikka acchan-aal pustakam kofakkapetfu
child-DAT father-INS book give-PASS-PST
'The book was given to the child by the father'
(ASHER & KUMARI 1997:208)
2.1.2 Ergatives
Similarly, an Ergative (ERG) is the grammatical realization given to the Agent in
the transitive construction of many languages. This falls within the so-called A
relationship, as proposed in DIXON (1994, 2000). As a basic illustration of the
category, please consider examples (3 .a) and (4.a) from Basque and Tokelau.
These examples contrast with (3.b) and (4.b), where the Subject of an intransitive
construction (the so-called S relationship) is given an unmarked encoding2:
Basque (LI)
(3 .a) Miren-e-k liburu bat irakur-r-i d-u-0
S> Mary-L-ERG book:ABS one:ABS read-L-PERF 3:ABS-WGW("have")-3:ERG
'Mary has read one book' (MANANDISE 1988:8)
1 The Passivizer Sf.-οί is restricted to passives with a third person Agent. (CRAIG 1977: 77).
2 In Basque, as in other ergative languages, the distribution of case marking is complex. There are
some intransitive verbs which encode their argument as A, resembling a semantically motivated ac-
tive pattern (CUTTING 1994). TRASK (1998), however, has pointed out that there are several verbs
which do not imply any volitionality or control.

30. 12 2. Defining the Object of Study
(3.b) Miren-0 ibill-i d-a
<$> Mary-ABS walk-PERF 3:ABS-/Z/W("be")
'Mary has walked' (MANANDISE 1988:8)
Tokelau (AU)
(4.a) ko te pahehe na totogi e te malo
TOP:TH SP fare PST pay ERG SP government
'The government paid the fare' (HOVDHAUGEN et a!. 1989:49)
(4.b) koi vili 0 te tama
CONT run ABS SP boy
'The boy is still running' (HOVDHAUGEN et al. 1989:85)
As examples (3 .a) and (4.a) show, I have only taken into consideration Ergative
marking applied to NP's, thus ignoring pronominal marking, which at times
receives a different treatment altogether in the grammatical apparatus of the
language. Similarly, the language in question may at times show an ergative split
concerning animacy (SILVERSTEIN 1976). When such a split is at issue, it is
commonly the case that ergative encoding applies only to NP's expressing entities
ranking low in the hierarchy; that is, entities which most often than not play a pas-
sive role in events. When this applies, the marker does not obviously refer to the
type of Agent profiled in § 2.1 above, but only to a marginal example of it.
Additionally, the encoding of NP's above illustrate a dependent-marking
pattern, as proposed in NICHOLS (1985, 1986, 1992). In this study, I have solely
taken into account Ergatives that behave in a dependent-marking fashion. The al-
ternative pattern, head-marking, is illustrated in (5) in Abkhaz (CA) with the use of
the prefix -y-; and in (6) in Basque (LI) with the suffix -te
(5) a-xac 'a a-pHas a-sq'a 0-1 θ-y-te-yt'
ART-man ART-woman ART-book (3SG:ABS)-3SG:DAT-3SG:ERG-gave-FINITE
'The man gave the woman the book'
(HEWITT 1979:36, from NICHOLS 1992:51)
(6) andereno-e-k ume-e-i ipuin asko-0 konta-tzen
teacher-DEF:PL-ERG child-DEF:PL-DAT story many-ABS tell-HAB
d-0-izk-0-i-e-te
3ABS-(PRES)-ABS:PL-i/A^Af("have")-DAT-3PL:DAT-3PL:ERG
'The (female) teachers tell many stories to the children'
(Modified from SALTARELLI 1988:149)
As shown by example (6), Basque illustrates a language with double marking,
given that both dependent and head-marked patterns apply. Head-marked Ergatives
constitute a closed class system -at times only devoted to cross-reference- and they
do not commonly show case-syncretism with any other semantic categories3. This
behavior contrasts with dependent-marked Ergatives, which often show syncretism
as will be shown shortly. For this reason head-marked Ergatives are not considered
3 Mayan languages are perhaps an exception.

31. 2.2 Agent marking NOT treated in the study 13
in the study, and only cases such as the suffix -k in Basque -examples (3 .a) and
(6)- or the preposition e in Tokelau -example (4.a)- will be taken into account.
2.2 Agent marking NOT treated in the study
One of the main goals of this study is to uncover the possible extensional paths that
may have led a number of markers to become grammatical markers of Agents.
Case-syncretism is taken here as the fundamental tool in the analysis. For this
precise reason, unmarked cases of Agents in transitive constructions, as typically
found in accusative languages, are completely uninformative for our purpose. The
same occurs with Nominative markers, which in most cases are only used to
encode a constituent as Subject, and nothing else. Neither case can inform us about
possible conceptual/semantic linkages with other categories. Compare the unmark-
ed case of Kannada in (7) with the oblique encoded Passive Agent illustrated in
(8)4, whose marker shows a wide case-syncretism as it can encode both an Instru-
ment (9) and the source location in an event of motion (10), (SRIDHAR 1990):
(7) akka-0 huDuganige haNa koTTaLu
sister-NOM boy-DAT money give-PST-3SG:F
'The sister gave the money to the boy' (p. 166)
(8) i: pustakavu vidva:-mantrigaL-inda biDugaDe
this book-ACC education-mirnster-/iV£M:PASS-AG realease
ma.DalpaTTitu
do-INF-experience-PST-3SG:N
'This book was released by the education minister' (p. 164)
(9) 5 'arjkara ko:lin-inda jhariyannu attida
Shankara stick-/MCM:INSTR centipede-ACC lift-PST-3SG:M
'Shankara lifted the centipede with a stick' (p. 168)
(10) ni.vu ya:va kaDev-inda baruttiddi.ril
you which side-/.V£M:ABL come-PROG-NPST-2HON
'Where are you coming from?' (p. 167)
The same phenomenon is illustrated in (11) in Japanese, where the Nominative
suffix -ga cannot encode other semantic categories in a syncretic fashion. This con-
trasts with the marking for Passive Agents in (12), where the suffix employed can
also express a location (13), as well as the Recipient of a transfer (also in 11):
(11) John-ga Mary-ni tegami-o okutta
John-NOM Mary-M:REC !etter-ACC sent
'John sent a letter to Mary' (TAKEZAWA 1993:45)
4 According to SRIDHAR (1990: 164), the specification of the Agent is not very common in the
language. This is, however, a rather common pragmatic phenomenon (SHIBATANI 1985).

32. 14 2. Defining the Object of Study
(12) banana-ga John-ni tabe-rare-ta
banana-NOM John-W/: PASS-AG eat-PASS-PST
Ά banana was eaten by John' (NISHIGAUCHI 1993:80)
(13) kooen-ni John-ga iru
park-M:LOC John-NOM is
'John is in the park' (TAKEZAWA 1993:55)
2.2.1 Other unmarked Agents
Occasionally, Passive Agents do not receive any overt marking. Such a strategy is,
nonetheless, rare across languages. Such a phenomenon appears in Imbabura
Quechua, as illustrated in examples in (14), (COLE 1985):
(14.a) Marya-ka [Juzi-0 ] riku-y tuku-rka
Maria-TOP Josi see-INF become-PST:3SG
'Maria was seen by Jose' (p. 107)
(14.b) [Juzi-ka] Marya-ta-mi juya-n
Jose-TOP Maria-ACC-VAL love-3SG
'Jose loves Maria' (p. 103)
Imbabura Quechua has a rigid SVO word order. The pragmatic demotion of the
Agent in the passive is achieved by means of thematizing the NP expressing the
Patient. The NP expressing the Agent does not receive any overt marking, while
the NP referring to the Patient is encoded as topic by means of the marker -ka.
Such a marker would apply otherwise to the Agent in the active counterpart, as
seen in (14.b), with the Patient then receiving the accusative marker -to. A similar
phenomenon is found in the Mayan language Tzeltal5, as examples in (15) show.
Example (15.a) is an instance of the common oblique encoding of a Passive Agent
in the language, while (15.b) shows an alternative unmarked version, (G. POLIAN
p.c.):
(15.a) mil-ot Juan [y-u'un te Pedro]
kill-PASS Juan 3:ERG-PASS-AG CL Pedro
'John has been killed by Pedro'
(15.b) mil-ot Juan [0 te Pedro]
kill-PASS Juan PASS-AG CL Pedro
'John has been killed by Pedro'
Cases such as these, though interesting, will not be treated in the analysis since
they do not display case-syncretism.
51 want to thank GILLES POLIAN for drawing my attention to these Tzeltal data.

33. 2.2.2 Active and Pragmatic Agents 15
2.2.2 Active and Pragmatic Agents
In the sample, I have only taken into consideration markers for Passive and
Ergative Agents. Active patterns present in active languages, where both A's and S-
A's are given marked realizations, are unfortunately not considered in the study.
Additionally, most active patterns are head-marked, and if dependent-marked, I
was unable to locate relevant data of such languages.
Similarly, it happens at times that certain agentive participants receive a special
encoding in certain contexts. The marking is not as conventional or stable as the
one commonly found in passive or transitive constructions, and it often applies for
pragmatic purposes. However, the pragmatic function at issue varies considerably
across languages; the markers are often used for either highlighting, contrasting,
topicalizing or even defocusing the Agent participant within a given predication.
Even though passive constructions have been treated as emerging from the appli-
cation of pragmatic functions (HOPPER & THOMPSON 1980, SHIBATANI
1985, etc.), their structure is rather conventional in grammatical terms. Addi-
tionally, it is somehow dubious that these pragmatic Agent markers represent any
unified category cross-linguistically, although further research into this area is
needed, and would be highly valuable.
If the use of such pragmatic markers becomes stable at any given point, the
markers in question may well become encoders of either Passive or Ergative
Agents in due time. In most cases, however, they are the only grammatical traits of
former productive patterns encoding Agents that have remained in marginal con-
texts. This is for example the case in Mising (ST), as illustrated in (16), where the
Agent receives a special coding when it is topic, (PRASADA SASTRY 1984:93):
(16) piyawe tabrochö-nä όά
bird:NOM hunter-AG:TOP shoots (unglossed)
'The hunter shoots the bird'
Another example of such markers is the use of the Ablative suffix -ngoonyi in
Jaminjun (AT), which is used to profile an unexpected Agent (SCHULTZE-
BERNDT 2000 calls this a "Contrastive Agent"). This is illustrated in (17), where a
kangaroo starts talking to two hunters to their complete surprise, (SCHULTZE-
BERNDT 1998: § 4.2.2.3: ex.89):
(17) " nanggayan goony-bi-yarlooga?"
who 3DU:3SG-FUT-poke
gani-yoo-boonyag, yangarra-ngoonyi-marlang
3SG:3SG-say/do:PST-3DU:OBL kangaroo-ABL-EMPH
' "Who do you want to spear?" it said to the two, the kangaroo did'
These pragmatic Agent markers are not taken into account in the sample because
they constitute an ill-defined category. Before proceeding with the analysis, I will
introduce the organization and data found in the samples in the next chapter.

34. 3. The Samples
The present study is based on the interpretation of the data found in two different
language samples. The first sample is the Sample of Agent Markers, to which I
often refer to as the Main Sample, or simply as the Sample. The other sample is the
General Sample. The first sample is fundamental for the type of analysis pursued
here, while the relevance of the second one is only secondary. In the process of
compiling the samples, languages were selected at random. Nevertheless, the scope
of my choice was frequently constrained by difficulties in the accessibility to the
data. Most of the data in the samples come from reference grammars, whose degree
of scientific quality vary greatly, but I have occasionally been able to obtain data of
some languages from native speakers or other language experts.
3.1 The Sample of Agent Markers or Main Sample
The Sample of Agent Markers was elaborated with the purpose of obtaining a
broad view of the typological behavior of case-syncretism of markers encoding
reference to an Agent. The Main Sample consists of markers for Agent found in
148 different languages. The relevant data are presented at the end of the book in
Appendix 1 (comprising both languages and references); Appendix 2 (Passive
Agent markers); and Appendix 3 (Ergative markers).
It is common that a given language may present alternative grammatical
markers to encode a certain NP as Agent within the same construction. For this
reason, the number of markers is expected to be higher than the number of
languages displaying them. In this light, the Sample has 176 Agent Markers in to-
tal, out of which 106 are for Passive Agent and 70 for Ergative Agent. There are
four languages in the Main Sample that allow the encoding of both a Passive and
an Ergative Agent, e.g. West Greenlandic (EA) where the Sf. -p encodes an Er-
gative while the markers -mit and -nut render an Agent in the passive. The relevant
figures are summarized in table (1):

35. 3.1 The Sample of Agent Markers or Main Sample 17
AG PASS-AG ERG
Languages 148 * 87 65
Agent Markers 176 106 70
Table 1: Number of languages and markers in the Main Sample
The asterisk serves as a reminder of the four languages that display Ergative
markers and markers for a Passive Agent. The distribution of these four languages
is significant since three pertain to the Eskimo-Aleutian macro-phylum (Siberian
Yupik, Alaskan Yupik and West Greenlandic) while the fourth is Athpare, a Ti-
beto-Burman language. This phenomenon should be treated as a family trait of
Eskimo-Aleutian, since we find it in three out of the five languages represented in
the Sample (there are actually six languages, but one was spoken in the nineteenth
century). If we accept that there are nine different languages in the family
(RUHLEN 1987:329), the representation of the family ranks up to 55.5 percent. In
contrast, the same phenomenon could be said to be rare within Tibeto-Burman,
since it only occurs in one out of the 12 languages attested in the Sample.
In table (2), I have arranged the languages of the Sample of Agent Markers
according to the macro-phylum they pertain to:
All AG PASS-AG ERG
Lgs. Mks. Lgs. Mks. Lgs. Mks.
Afro-Asiatic AF 6 7 6 7 -
Altaic AL 4 7 4 7 - -
Amerind1 AM 19 19 13 13 6 6
Australian AT 25 29 5 7 20 22
Austro-Asiatic AA 1 1 1 1 - -
Austronesian AU 11 13 7 9 4 4
Caucasian CA 12 13 1 1 11 12
Chukotko-Kamchatkan CK 2 3 1 1 1 2
Elamo-Dravidian ED 3 3 3
*>
»
- -
Eskimo-Aleutian EA 6 11 6 8 3 3
Indo-European IE 26 34 23 30 3 4
Indo-Pacific IP 3 3 - - 3 3
Niger-Kordofanian NK 4 4 4 4 . -
Nilo-Saharan NS 1 1 1 1 - -
Sino-Tibetan ST 14 16 4 5 11 11
Uralo-Yukaghir UY 4 5 3 4 1 1
Language-Isolates: LI
Ainu 1 1 1 1 - -
Basque 1 1 - - 1 1
Japanese 1 1 1 1 - -
Korean 1 1 1 1 - -
Tarascan 1 1 1 1 - -
Trumai 1 1 - - 1 1
Warao 1 1 1 1 - -
TOTAL 148 176 87 106 65 70
Table 2: Number of languages and markers by macro-phyla
1 The group "Amerind" is not taken to represent any macro-phylum per se. It is intended as a con-
venient cluster of different linguistic families spoken in America.

36. 18 3. The Samples
One of the weaknesses of the Sample resides in the fact that Indo-European and
Australian have a greater representation than desirable. On the other hand, the pres-
ence in the Sample of macro-phyla, such as Niger-Kordofanian or Indo-Pacific, is
almost irrelevant considering how large these families are. Nevertheless, many of
the deficiencies of the Main Sample are due to the very nature of the object of
study.
As for Passive Agents, one must bear in mind that not all languages have
passives. In fact, it seems that a full-fledged West Indo-European-like passive is
not that common typologically when we look closely at the world's languages.
Among languages that have a passive, only a few allow the specification of the
Agent. These factors greatly complicate the search. Such phenomena may even be
characteristic of an entire group or type; many languages of New Guinea are
passive-less. The same can be said of Chadic languages, (Hausa being an ex-
ception, as KEENAN (1986) points out). The same happens in Khoisan, in Na-
Dene, together with the vast majority of languages spoken in North and Central
America (i.e. Oto-Mangue), and most Creoles, if not all. Nilo-Saharan languages
have agent-less passives (Luo is the only language of this family represented in the
Sample. Even here, what I take to be an Agent marker is a very recent, prag-
matically marked innovation from the contact with Hausa). Other examples of
agent-less passives within the macro-phylum are Classical Arabic and Tamazight.
A similar phenomenon applies for Ergative marking because not all Ergative
languages in the world encode ergativity on a dependent-marked basis. Since head-
marked ergativity is left out here as an object of study, the Sample of Ergative
markers lacks data from the Mayan family, the Mixean family, as well as many
Caucasian languages, especially the northwestern branch.
The languages in the Sample can also be grouped if we consider their
continental location. The results of such a sorting are displayed in table (3), where I
have taken into consideration six large geographical areas, mainly Africa; Europe
(West to the Ural Mountains); Eurasia (comprising the Caucasus, the Middle East
and all Asia); Oceania (comprising all islands in the Pacific); Australia and
America.
PASS-AG ERG
Lgs. % Lgs. %
Africa 8 9.1 0 0.0
Europe 19 21.8 1 1.7
Eurasia 27 31.1 27 41.5
Oceania 7 5.8 7 10.7
Australia 5 8.1 20 30.7
America 21 24.1 10 15.4
Total 87 100.0 65 100.0
Table 3: Languages by continent

37. 3.1 The Sample of Agent Markers or Main Sample 19
The distribution of these geographical areas by Agent type is presented by way of a
graphic in chart (1):
Chart 1: Languages by geographical area
As we can see in chart (1), most markers for Passive Agent in the Sample come
from European, Eurasian, and American languages. The presence of African or
Oceanic languages is minor. As for Ergative markers, Eurasian and Australian lan-
guages heavily represent the Sample. It is expected that African and European
languages do not play any significant role here given that ergativity is an (almost)
absent grammatical phenomenon in these geographical areas, (save for Basque,
whose presence gives Europe the value of a 1.7 percent).
The data in table (3) may alternatively be observed from a different perspective.
We may now observe the distribution each type of Agent displays within each
geographical area. The contrasts that result are presented graphically in chart (2).
As expected, all markers from African languages in the Sample are of a Passive
Agent. The same applies for 95 percent of the markers from European languages.
The distribution is well balanced in Eurasian and Oceanic languages. However,
Australian languages show a preference for ergative organization, which contrasts
with languages from Europe. As for languages from the Americas, there are more
Passive Agents than Ergatives, although ergativity is relatively common in South-
ern Meso-America and in the Amazon, being less common (it seems) in other
parts, especially in North America (where the active pattern appears pervasive).

38. 20 3. The Samples
100
80 -
60 · I
40 -
20 - .||j
0 4—1 1 . L 1—r—' — 1 I I • — —
Africa Europe Eurasia Oeania Australia America
• PASS-AO Ο ERG
Chart 2: Agent types by geographical area
3.2 The General Sample
The General Sample serves as an auxiliary database for the Sample of Agent
Markers. This sample consists of 215 markers that encode semantic categories such
as Instrument and/or Cause. NONE of the markers in this sample has undergone an
agentive extension. The markers of the General Sample come from 162 different
languages, 70 out of which provide data on Agents in the Main Sample. In table
(4), the languages and markers of the General Sample have been arranged by
macro-phylum:
Macro-Phylum LRS. Mks. Macro-Phylum Lgs. Mks.
Austro-Asiatic AA 1 2 Na-Deni ND 2 4
Afro-Asiatic AF 7 9 Niger-Kordofanian NK 6 6
Altaic AL 6 9 Nilo-Saharan NS 6 7
Amerind AM 42 50 Sino-Tibetan ST 6 11
Australian AT 13 17 Uralo-Yukaghir UY 4 5
Austronesian AU 13 15 Language Isolates:
Pidgins & Creoles C 4 6 Ainu 1 2
Caucasian CA 10 15 Basque 1 3
Chukotko-Kamchatkan CK 1 1 Japanese 1 1
Elamo-Dravidian ED 4 5 Korean 1 1
Eskimo-Aleutian EA 3 3 Tarascan 1 1
Indo-European IE 21 31 Warao 1 2
Indo-Pacific IP 4 6 Wari' 1 2
Khoisan KH 1 1 I Total 162 215
Table 4: Languages and markers in the General Sample by macro-phyla

39. 3.2 The General Sample 21
All these languages together with their corresponding markers are presented in
Appendix 4. There are 92 languages that are not represented in the Sample of
Agent Markers. Their absence in the Sample is motivated by at least one of the
following reasons:
- They may not have a passive at all.
- They may have an agent-less passive (save for Catalan and French).
- If ergative, they do not encode the A function of a given NP on a dependent-
marked fashion.
In the next chapter, I will introduce the Agent markers in the Main Sample that
show "asyncretism". In other words, we will see specific markers for Agents,
which do not syncretize with any other relevant category.

40. 4. Asyncretic Agent Markers
Among the 176 Agent markers in the Main Sample, some do not show any
apparent case-syncretism with other relevant semantic categories. I will treat such
cases as "asyncretic". The exact figures are given in table (1):
All AG PASS-AG ERG
Mks. Mks. % Mks. %
Asyncretic 22 8 7.6 14 20.0
Syncretic 154 98 92.4 56 80.0
Total 176 106 100.0 70 100.0
Table 1: Ratio of syncretism in Agent markers
There is an immediate difference here between markers for Passive and Ergative
Agents concerning the degree of asyncretism. 92.4 percent of Passive Agent
markers display some type of syncretism. The percentage left for asyncretism does
not even reach the eight percent. This contrasts with the degree of asyncretism of
Ergative markers, which reaches up to 20 percent. These figures may be taken to
point to a different degree of grammaticalization for the markers involved, the
degree being much higher for Ergative markers than for markers for Passive Agent.
In the prototypical case, the coding of an Agent in a passive construction is reg-
ulated by pragmatic purposes, and in many languages, its instantiation is not even
allowed. In contrast, this pragmatic character is very rare when coding an Agent in
the transitive construction, which is often ruled by semantic or syntactic purposes.
This makes Ergative markers more prone to grammaticalization because their func-
tion specializes further, first as semantic markers of Agent and then as syntactic
encoders of (transitive) Subject. Such grammatical constraints may put pressure on
the semantics of the marker involved to trigger the gradual loss of its semantic
transparency. This in turn may affect its links to other categories. All these
processes lead to morphological isolation, which comes together with the increase
of functional specification, resulting in asyncretism.

41. 4.1 Further contrasts: One or more markers to express Agent 23
4.1 Further contrasts: One or more markers to express Agent
The same phenomenon affecting the asyncretism of Ergative and Passive Agent
markers motivates a further difference between them. This difference concerns the
extent to which the language has one or more options to encode such a participant.
For Passive Agents, it is often the case that the language has more than one marker
available to encode the NP expressing the Agent. This variation is free in the typi-
cal case, that is, the speaker opts for one or another marker for pragmatic or
stylistic reasons. When there are alternative markers for a Passive Agent, it is
always the case that the markers in question have a different syncretism.
In contrast, ergative languages often have only one Ergative marker to encode a
given NP in A function. Where various markers are present, it is typically the case
that they are allomorphs, and that their distribution is conditioned by the language
(i.e. lexical, phonemic, etc.), and not by the speaker. At times, the variants may
display a different syncretism even though they might stand as conditioned al-
lomorphs. When this is the case, I have considered them as different markers.
Allomorphs with the same syncretism are seen as simple variants of the same
Ergative morpheme. For example, Brokskat (IE) has three Ergative allomorphs
which are grammatically conditioned: the Sf. -ya (applies to nouns and pronouns in
the past tense); the Sf. -/ (applies to Proper Nouns in the past tense); and the Sf. -sa
(applies in non-past). The first one is asyncretic, while the other two show syn-
cretism with an Instrument. In this sense, I will take Brokskat to display only two
Ergative markers, one asyncretic and the other syncretic, even though the latter
consists of two allomorphs. Table (2) illustrates the contrasts among the markers in
the Sample:
PASS-AG % Mks.
71 lgs. with 1 mk. =81.6 71
16 lgs. with + than 1 mks. = 18.4
~~ ______ ^ 13 lgs. with 2 mks. 81.2 26
^—3 lgs. with 3 mks. 18.8 9
Total 87 lgs. 100.0 106
ERG % mks.
60 lgs. with 1 mk. = 92.3 60
5 lgs. with + than 1 mks. = 7.7
^ 4 lgs. with 2 mks. 80.0 8
1 lgs. with 3 mks. 20.0 3
Total 65 lgs. 100.0 71
Table 2: Ratio of Agent markers by languages
The contrast obtained is remarkable. Almost 19 percent of the languages of the
Sample have more than one option to encode a Passive Agent. The possibility of
the existence of a different marker with another syncretism is much lower for
Ergatives with less than eight percent. As for the number of alternatives, languages
agree here for both types of Agent markers, as two alternative markers are more
common than three (in a ratio of 80 against 20 percent).

42. 24 4. Asyncretic Markers
4.2 Languages and markers
The following table presents all asyncretic markers in the Main Sample organized
by language and macro-phyla:
Asvncretic PASS-AG's Asvncretic ERG's
AF Hebrew P. bi-dey AM Pemon (CR) Sf. -da {-yd)
AL Turkish PP. taraf-in-dan Uwa(CH) Sf. -t
AU Maori P. e AT Djabugay Sf. -njdja
IE Persian P. be dceste Warrwa Sf. -na / -nma
P. be tcEviEsote Sf. -yunu-na1
Spanish P. por parte de CA Bezhedukh-Adyghe Sf. -m
LI Warao PP. aisia Chechen-Ingush Sf. -(c)uo
UY Finnish PP. (GEN)-n toimesta Tsova-Tush Si.-s2
IE Brokskat Sf. -ya
ST Balti Sf. -si
Mizo CL. -in
Purki Sf. -is
LI Basque CL. -{e)k
LI Trumai Sf. {-*}
Table 3: Asyncretic markers
Ergative markers here are morphologically simple in practically all cases. The
exception is the Sf. -yunu-na in Warrwa, which also serves as an encoder of a
Pragmatic Agent (see § 2.2.2). All asyncretic Passive Agent markers, in contrast,
are complex both phonemically and morphologically. The exception here is the
particle e in Maori (Polynesian), which is the same morpheme used as Ergative in
other languages of the family, i.e. Samoan. As for the rest, most of them are com-
posite expressions of the type "at the hands of AG", "by the mediation of AG",
"through the part of AG", "at AG's occupation". I lack data to make a proper mor-
phological analysis of the case in Warao, but I can conclude at this point that true
asyncretic Passive Agent markers are not expected unless they occur as highly
specialized phrasal combinations that have a lexical (not grammatical) meta-
phorical meaning.
4.3 Summary
In this chapter, we have revisited the Sample paying special attention to those
Agent markers that do not show syncretism with any other category. Certain
relevant contrasts have been found when comparing markers for Passive and
Ergative Agents.
' Warrwa is taken here to have two Ergative morphemes because these markers are not allomorphs to
each other, but free variants.
2 The Ergative marker is also applied to a defocalized Agent in relative clauses where the Patient is
Subject, (HOLINSKY & GAGUA 1989).

43. 4.3 Summary 25
The first contrast concerns the ratio of asyncretism, which is much higher in
Ergatives than in Passive Agents. The second contrast concerns the number of
different markers available within the same language to encode the same
participant. As expected, the ratio is much higher in Passive Agents than in
Ergatives, as the choice of the latter is constrained by the grammar. Finally, we
have also seen in § 4.2 that asyncretic markers in Ergative languages are typically
monomorphemic, while asyncretic Passive Agents are complex morphemic
expressions whose (lexical) meaning is often transparent. In the next chapter, I will
survey the Sample again, but this time by focusing on syncretic markers. I will
compare the different categories involved in the syncretisms of both types of
Agents. The contrasts obtained are also interesting.

44. 5 Syncretic Agent Markers
As we saw in the previous chapter, Agent markers can be divided according to
whether they syncretize or not with a given relevant semantic category. Table (1)
from Chapter 4 is repeated here for convenience as table (1):
All AG PASS-AG ERG
Mks. Mks. % Mks. %
Asyncretic 22 8 7.6 14 20.0
Syncretic 154 98 92.4 56 80.0
Total 176 106 100.0 70 100.0
Table 1: Syncretic Agent markers
The figures in the table inform us that syncretism is a rather common phenomenon
in the markers conveying an Agent. Considering the type of asyncretic cases of
Passive Agents seen in § 4.2, I am now in a position to claim that a given marker
for Passive Agent is almost always expected to show some case-syncretism with
other categories, unless it is a complex expression. As for Ergative markers, the sit-
uation is less predictable, since monomorphemic asyncretic markers also abound.
In the following section, I will be presenting a brief description of the categories
involved in the syncretisms of the markers in the Main Sample. In § 5.3, the
syncretic patterns are associated, and the involvement of each category is presented
in order to understand the extent to which the categories involved display typo-
logical trends.
5.1 Categories involved in the syncretisms
In this section, I present the different semantic categories present in the encoding
possibilities of the Agent markers in the Sample. They are here divided as Primary
and Secondary according to the relevance and the frequency of their involvement:

45. 5.1.1 Cause 27
Thematic Spatial
Primary Cause (CAUSE) §5.1.1 Ablative (ABL) §5.1.7
Causee (CAUSEE) §5.1.2 Allative (ALL) §5.1.8
Dative (DAT) §5.1.3 Locative (LOC) §5.1.9
Intermediary (INTER) §5.1.4 Perlative (PER) §5.1.10
Instrument (INSTR) §5.1.5
Possessor (POSS) §5.1.6
Secondary Beneficiary
Comitative
(BENE)
(COM)
§ 5.1.U.a
§ 5.1.1l.b
Proximative (PROX) § 5.1.11.C
Table 2: Categories involved in the syncretisms.
Before proceeding further, a word about the methodology I applied when grouping
the sampling is in order:
"A marker X is taken to express a given semantic value Y, whenever X is
employed in the language to encode reference to Y, regardless of whether X is
or is not the most frequent means in the language to encode Y"
In this light, when a marker encodes reference to a given category, it is taken to
mean that the category is present in some degree in the spectrum of the encoding
possibilities of the marker in question. Similarly, the marker will be taken to
display a certain range of "semanticity" which treats the category as somehow re-
lated to other values that the marker may also express.
With this in mind, I will now proceed to present a brief characterization of the
categories in table (2). This characterization is again based on standard linguistic
theory and it is intended to present working definitions for the categories involved
in the syncretisms to bear in mind the semantic scenarios that define them.
Accordingly, the content of this section will be common knowledge for many read-
ers. Since a full reading may well turn to be superfluous for many, I recommend it
only as reference.
For the sake of convenience, I illustrate the categories by giving examples in
English, but some examples will be rendered in Spanish too. Further examples
showing case-syncretism with Agents are also provided.
5.1.1 Cause
A given marker is taken to express Cause (CAUSE) when the marking in question
refers to the role played by a given entity -normally conceived of as either an
abstract or natural force- construed as the causal force which has brought about a
certain state of affairs. In the prototypical case, the grammatical frame, where
Cause is instantiated, corresponds to an intransitive construction where the verb
and its core argument designates the state of affairs implied, and where Cause is
encoded as an Oblique non-argumental participant. Cause is illustrated in (1) with
the role played by the cold
(1) the woman diedfrom the cold

46. 28 5. Syncretic Agent Markers
The clause in (1) is intransitive. Here, a given state of affairs is predicated where a
woman is profiled to have undergone a change of state, e.g. from being alive to
being dead. In this predication, we are further informed that the presence of certain
extreme atmospheric conditions has triggered such an outcome. Here the NP
expressing Cause is encoded obliquely within an adpositional phrase whose head is
the preposition from. In Modern English, the PASS-AG is no longer encoded by
means of this preposition. However, this was one of the available coding
possibilities in Old English, where the marker in question - fram/from, depending
on the dialect - was also commonly used to express Cause. The former case is
illustrated in (1), taken from PALANCAR (1997a: 137):
(2) he weard ofslagen from his agn-um monn-um
he became slain FÄOMPASS-AG his own-DAT:PL man-DAT:PL
'He was slain by his own men'
In the Main Sample, we also find markers encoding both an Ergative Agent and
Cause, as in a syncretism of the type <CAUSE:ERG>. This type of syncretism is
found for example in Siberian Yupik as illustrated in (3), where the causal value is
applied to an atmospheric circumstance, and in (4) where we encounter the marker
encoding the A relationship in an instance of the transitive construction, (DE
REUSE 1994):
(3) [aghyum naavuk]
aghvugh-m naave-uk
get:old-M4:CAUSE break-IND:3SG
'It (a stove) broke because it's cold' (p.36)
(4) [qikmima neghaa kayu]
qikmigh-ma negh-aa kayu-0
dog-AM.ERG eat-IND:3SG-3SG fish-ABS
'My dog ate the fish' (p.30)
There is a relevant sub-case of Cause, which I have treated under the label
"Cause/Reason". Cause/Reason is the role of a participant construed as the reason
or the motive for which the action leading to the event takes place. As it will be
seen in more detail in § 9.3, the role of such participants can be both conceptually
construed as the Reason motivating the Agent to act and as the ultimate Cause
which triggered the event as a whole. Cause/Reason is commonly expressed in
English by means of the complex prepositions because of or due to, but the case is
perhaps better illustrated in Spanish with the use of the preposition por, as in (5):
(5) Juan se qued-ό en casa por el mal tiempo
John 3:R stay-3SG:PST LOC house POÄ:CAUSE the:M bad weather
'John stayed home because of the bad weather'
In (5), the bad weather is conceived of as both the Reason leading John to decide
staying at home, and the Cause why he did not go out. The same marker is pro-
ductively used in Spanish to encode causal events in subordinated infinitival

47. 5.1.2 Causee 29
clauses, as illustrated in (6). The marker has grammaticalized, together with the
general subordinator que, and has become the most common causal conjunction in
the language. This is illustrated in (7):
(6) Pedro se desmay-ό por no hab-er com-i-do
Peter 3:R faint-3:SG:PST TOÄ:CAUSE NEG have-INF eat-ThV-P.P
'Peter fainted because he hadn't eaten'
(7) Ana se fue a casa por-que en gl bar hac-ia frio
Ana 3:R went:3SG to house POÄ:CAUSE-SUB LOC the:M bar do-3SG:IMP cold
'Ana went home because it was cold in the bar'
This marker serves to illustrate the syncretism <CAUSE:PASS-AG>, as shown in
example (8):
(8) ese asunto fue estudi-a-do por especialista-s
that issue was:3SG study-ThV-P.P POR:PASS-AG specialist-PL
'That issue was studied by specialists'
5.1.2 Causee
The Causee (CAUSEE) is a complex role that corresponds to a subordinated Agent
in a causative construction. In these constructions, the Causee acts as instigated or
forced to do so by a higher Agent, whose role is commonly regarded as Causer. An
example of a causative scene is illustrated in examples (9) and (10) which
instantiate slightly different constructions existing in English for the purpose. In
both cases, John plays the role of a Causee:
(9) Mary made/had John buy a new car
(10) Mary had a new car bought by John
These examples have slightly different semantics. In the first case, John gets
involved in the action of purchasing a car, but he does not do it from his own
initiative, but because of the psychological force that he undergoes from his social
interaction with Mary. The clause also expresses that the whole event was initiated
because of Mary's volition. In the second example a similar situation applies, but
there is an additional beneficiary nuance on the role played by Mary, who this time
is construed as clearly benefiting from the action carried out by John. In example
(9) the Causee, John, is encoded as a primary object, while in (10) it is encoded in
oblique terms. The latter example serves as an illustration of the syncretism
<CAUSEE:PASS-AG>, as the Passive Agent is well known to be encoded in
English by means of the preposition by, as in (11)
(11) the car was bought by a strange couple
The syncretism <CAUSEE:ERG> is also found in the Sample. The Carib language
Tiriyo illustrates such a combination in (12), where both NP's can receive an

48. 30 5. Syncretic Agent Markers
ambiguous interpretation as either Causees or Agents. Tiriyo has an ergative split
based on tense, which applies in the so-called Narrative Past (MEIRA 1999:554):
(12) Sipaki t'i-tuuka-po-e pahko-ja karaiwa-ja
Spike NPST-hit-CAUSE-NPST 1 :father-J/f:CAUSEE/ERG Brazilian-^:CAUSEE/ERG
(i) 'Mv father made the Brazilian hit Spike'
(ii) 'The Brazilian made my father hit Spike'
5.1.3 Dative
Under the general rubric of Dative (DAT), I refer to a number of thematic roles that
are often encoded by means of Dative markers. The most typical role associated
with such a marking is the Receptor of an event of transfer, which depicts the
participant at whom the transfer of a given entity is aimed. Recipients are most
commonly encoded in English as primary objects in a ditransitive construction, as
in (13), but they can also receive an oblique encoding at times, as in (14):
(13) Mary gave John a present
(14) Mary gave the present to John
Since the preposition to is not used in English to encode a Passive Agent, this
language cannot serve as illustration for a syncretism <DAT:PASS-AG>. Such a
pattern is, nonetheless, found in many other languages; Korean illustrates it in
example (15) and (16) with the morpheme {-ey}, (SOHN 1994)1:
(15) minca-nun namwu-ev mwul-uI cwu-ess-e
Minca-TOP tree-£K:DAT water-ACC give-PST-INF
'Minca gave water top the trees' (p.238)
(16) tokwuk-i swunkyeng-evkey cap-hi-ess-ta
thief-NOM police-£T:PASS-AG catch-PASS-PST-DC
'The thief was caught by the police' (p.242)
Tiriyo serves again as an illustration for a syncretism <DAT:ERG>. Consider
examples (17) and (18) for this purpose, (MEIRA 1999):
(17) sen wi'-r'i ee-ja
3:INPROX 1 :ERG-make:PST:PERF l-JA-.ΌΚΊ
Ί made this for you' (p.424)
(18) emeine t-epoo-se ii-ja
thorn NPST-find-NPST 3-J/f:ERG
'S/he found thorns' (p.425)
1 Examples illustrate the allomorphs -ey and -eykey. The former is used with nouns of inanimate
referents (namwu- "tree"), the latter with animates (swunkyeng- "police").

49. 5.1.4 Intermediary 31
5.1.4 Intermediary
The Intermediary (INTER) corresponds to the role played by an agentive
participant who facilitates that another participant succeeds in carrying out a given
event. Example (19) illustrates the category:
(19) Mary got to know a lot of people through John
The role played by John in (19) is of an Intermediary. He is construed as having
facilitated that Mary met many people by means of a series of unspecified actions
(e.g. getting people interested in Mary, getting them together, introducing people to
Mary, etc.). Here the Intermediary is encoded with the preposition through, but it
could equally be realized by other means such as thanks to, by means of, or simply
with the use of via. None of these encodings can be used to express the Agent in a
passive construction in English.
The syncretism between Intermediaries and Passive Agents can be illustrated in
Spanish again with the use ofpor. As English, Spanish may alternatively make use
of the preposition a troves de "through" to express the Intermediary. This is shown
in the following examples:
(20) Maria conoci-ό a much-α gente a travis de /por Juan
Mary meet-3:SG:PST OBJ many-F people through John
'Mary got to know a lot of people through John'
The Intermediary is encoded in (20) by means of two possible markers. If the
marker por is used, the sentence has an ambiguous reading concerning the role
played by the referent of the NP Juan (cf. PALANCAR 1996:78-79). The role of
John can alternatively be construed either as the Intermediary or as the Beneficiary
of the event. However, one reading excludes the other; that is, they are asym-
metrical, and do not emerge through a semantic continuum. In the Benefactive
reading, Mary gets involved in meeting many people so as to benefit John in some
unspecified way with the out-come of such social interactions. Additionally, the
role of John may alternatively be construed as playing the role of an Indirect
Causer. Under this reading, Mary is construed as getting involved in this social
interaction because John has convinced her to do so.
All these readings are not fortuitous, and they can be accounted for to some
degree as resulting from the wide syncretism present in the semantics of the
preposition por. Up to now, the marker renders the following pattern: <CAUSE:
CAUSE/REASON:CAUSEE:INTER:PASS-AG>. To these readings, we may add
the Beneficiary (see § 5.1.1 l.a below). Nevertheless, such an account is only
satisfactory under a descriptive perspective, for it does not inform us why such a
syncretism applies and not other.
For the time being, I will take a mere descriptive position upon such a phe-
nomenon. This position allows us to approach the data with little theoretical bias.
At this first stage in the analysis, I have assumed with no further pondering that
case-syncretism occurs in the languages of the world as a fact. Under such a view,

50. 32 5. Syncretic Agent Markers
the categories I am presenting in this section are seen as isolated cases that happen
to occur in combination with others. In the remaining parts of this book, I will take
a rather different perspective.
In contrast to Passive Agents, NO marker in the Sample illustrates the syncre-
tism of Intermediary and Ergative.
5.1.5 Instrument
The Instrument (INSTR) is defined within a prototypical instrumental scene. I will
entirely devote Chapter 8 to achieve a better characterization of this role. The
Instrument is the role played by the object the Agent manipulates to achieve a
change of state on the Patient. Such a change of state is perceptually clear, and re-
sults from the direct force-dynamic impingement of the Instrument on the Patient.
The Instrument is illustrated by the role played by the knife in the event portrayed
in (21):
(21) Pete sliced the bread with a sharp knife
Here the Agent, Pete, manipulates the knife to effect a sliced change of state of the
bread. The Instrument in (21) is grammatically realized in oblique terms by means
of the preposition with. Since English does not make use of this marker to encode a
Passive Agent, this language cannot illustrate the syncretism2. For this purpose,
consider examples (22) and (23) which show the syncretism <INSTR:PASS-AG>
in Jacaltec (MY:AM), (CRAIG 1977):
(22) x-0-in-mak metx tx'i [y-w hune? te?
ASP-3SG:ABS-lSG:ERG-hit CL dog 3:ERG-t/:INSTR CL:a stick
Ί hit the dog with a stick' (p. 17)
(23) x-0-mak-ot naj pel \y-u mi xuwan]
ASP-3SG:ABS-hit-PASS CL Peter 3:ERG-C/:PASS-AG CL John
'Peter was hit by (because of) John' (p.78)
Similarly, take example (24) from Avar (TCHEKHOFF 1979:155), which shows
the pattern <INSTR:ERG>:
(24) di-ca ost?o-ca c?ul q?ot?-ana
& 1SG-C4:ERG axe-C/i: INSTR wood cut-PST
'J'ai coupe du bois avec la hache' / Ί cut wood with the axe'
2 Some may want to argue that por conveys instrumental notions, as it can be used for the expression
of some means of communication, such as the phone in por telefono "by phone". However, such
notions are marginal to the Instrument prototype, which por cannot express, i.e. * Maria corto el pan
por un cuchillo with the intended meaning "Mary cut the bread with a knife".

51. 5.1.6 Possessor 33
5.1.6 Possessor
The Possessor (POSS) is the entity in control over another entity (the Possessum),
which lies within the accessibility domain of the Possessor. The Possessor enters in
a part-whole relationship with the Possessum, being the latter construed as a part of
the former. In grammatical terms, the Possessor is often encoded as a complement
of the NP expressing the Possessum. As LANGACKER (1993) has pointed out,
Possessors serve as optimal Reference Points for the cognitive accessibility of
certain entities. The Possessor is expressed in English via two constructions: the
Genitive case in (25), and the Oblique o/-phrase in (26)3. These two frames
alternatively bear slightly different distributions:
(25) Mary's car was once red
(26) the light of the Sun
The latter encoding served in Medieval English as the most common encoder of an
Agent in the passive construction. This option is shown in example (27), (PA-
LANCAR 1997a: 137):
(27) pe puple trow-ip betere perto whanne it is seyd
the people believe-3:PL better thus when it is said
of a maistir
OF:PASS-AG a master
'People believe it more so when it is said by a master'
Similarly, the combination <POSS:ERG> is also found across languages. Limbu, a
Tibetan language, illustrates the case in examples (28) and (29), (VAN DRIEM
1987):
(28) khune? tumma-re ku-sa?
he first:wife-ÄE:GEN her-child
'He's the son of (our father's) first wife' (p.43)
Φ lit. 'He (is) her child of the first wife'
(29) mgna-re menda?-?en me-se?r-u
man-«£:ERG(:PL) goat-ABS NSGA-kilI-3PL
'The men killed the goat' (p.40)
5.1.7 Ablative
The term Ablative (ABL) is used here to refer to the source location of an event of
motion. In the spatial scenario that defines the category, the Ablative is the ground
with respect to which the figure effects the motion event, as it moves in such a way
that it progressively weakens its spatial contact with the ground. When I gathered
3 Alternatively, there are other construals of possession relationships such as, the key to the door, or
the entrance to the hospital.

52. 34 5. Syncretic Agent Markers
the data in the Main Sample -as pointed out in § 5.1 above, I considered that a
given marker was able to express an Ablative value if there was enough evidence
that the marker in question could express a motion spatial scenario similar to the
one described. Nonetheless, languages are often very sensitive to other factors,
such as for example the type of ground involved, or the type of path described; and
they may use different markers accordingly. The case is illustrated in Spanish and
English in example (29):
(29)
a. el escalador salt-ό d-el pico mäs alto
the:M climber jump-3SG:PST D£-the:M peak more high
b. the climber jumpedfrom the highest peak
In these examples, the climber (as figure) effects an event of motion with respect to
the peak (as ground). In this event, the peak is understood as the source location
where from the climber moves. In the Spanish example, the marker used to convey
the Ablative scenario does not specify any type of ground configuration, and the
same can be said for English. However, English can also use another marker when
the ground is construed as a surface location, so that the figure and the ground are
conceived of as conforming some unity due to close contact. The stability of such a
contact is altered in the event of motion. This is shown by off in example (30).
Similarly, off could also apply to example (29.b) and it would highlight the
detachment of the climber from the surface of the peak:
(30) the paper came off the wall
In Spanish, another Ablative preposition, desde, equally insensitive to the type of
ground, expresses an anchored trajectory initiated in the ground. The marker may
additionally render an inchoative or a contrastive informative nuance. The case is
illustrated in (31), where the ground is portrayed as being a salient landmark with
respect to other possible alternatives:
(31) Juan se fue and-a-ndo a-l cine desde su casa
John 3:R go:3SG:PST walk-ThV-GER to-the:M cinema DESDE 3:POSS house
'John walked to the cinema from his place'
I do not want to show with these examples how Spanish or English construe spatial
relations in linguistic terms, rather, I wanted to illustrate that within a given
language there may be alternative markers at issue to construe a given spatial scene
in a different way. The markers de, desde, from and off, are treated here as cases of
Ablative, regardless of their semantic specification. The same applies to other lan-
guages in the Sample. As seen in example (2) above, the marker from can illustrate
the combination <ABL-PASS-AG> in Old English. Similarly, the preposition off
would exemplify the same syncretism for Medieval English, as shown in (27)
above, given that modern off and of were still the same marker at this stage of the
language. Similarly, the preposition de was the most common encoder of Passive
Agents in Old Spanish. Its use for this function is rather marginal today, but it
survives when certain verbs are used. This is illustrated in (32) and (33):

53. 5.1.8 Allative 35
(32) Pedro i-ba acompan-a-do de su madre
Peter go-3S:IMP accompany-ThV-P.P £>£:PASS-AG 3:POSS mother
'Peter was accompanied by his mother'
(33) la-s zona-s pantanos-a-s suel-en est-ar
the:F-PL zone-PL marshy-F-PL be:often-3PL be:located-INF
infest-a-d-a-s de mosquito-s
infest-ThV-P.P.-F-PL Z)£:PASS-AG mosquito-PL
'Marshy areas often thrive with mosquitoes'
lit. 'Marshy areas are often plagued by mosquitoes'
The combination <ABL-ERG> is also found in the Sample, and is illustrated in
examples (34) and (35) from Tauya (IP), (MACDONALD 1990):
(34) yau?a-ra 0-neme-ni-?a: i?ai-ra 0-fateifo-ni-?a
yam-TOP 3SG-head-N/:ABL-IND yam-TOP 3SG-shin-A7 ABL-IND
'(After she died), yau?a-yams (emerged) from her head; i?ai-yams from
her shins' (p.318)
(35) ?e fanu-ni fena?a-ra 0-yau-a-?a
DEM man-MERG woman-TOP 3SG-see-3SG-IND
'The man saw the woman' (p.321)
5.1.8 Allative
The Allative (ALL) is again the generic label given to a spatial scenario where the
figure describes a path directed to a given Ground. The marker that most typically
realizes this type of scenario in English is the preposition to, as shown in (36), but
toward(s) would equally serve the purpose:
(3 6) we were strolling along the street to the harbor
English again cannot serve to illustrate any Passive Agent combination. The case is
illustrated in (37) and (38) in Nigerian Arabic (AF) (OWENS 1993):
(37) fi yerwa masee-t le biet rafiig-i
LOC Yerwa went-lSG ££:ALL house friend-lSG:POSS
'In Maiduguri (sic.) I went to my friend's house' (p. 158)
(38) al-äkil ma bin'dkil le-i
DEF-food not eaten L£.PASS-AG-me
'This food can't be eaten by me' (p. 157)
Similarly, Tiriyo illustrates the syncretism <ALL-ERG>. For this purpose, com-
pare example (18) above with (39), (MEIRA 1999: 424):
(38) ire-npe-pee Paumira ki'n-te Mirike-ja
3: IN AN A-PST-ABL Paumira 3:PST:PERF-go Mirlke-^:ALL
'After that Paumira went to Mirike' (p. 157)

54. 36 5. Syncretic Agent Markers
5.1.9 Locative
The Locative (LOC) refers here to the spatial scenario where a given entity, being
an object or an event, is located with respect to a ground, this being a larger object
or a place. In the prototypical case, the Locative refers to a static scenario. Lan-
guages differ considerably in the degree to which they are sensitive to other more
specific configurations within this general picture, i.e. geometrical shape of the
figure, posture, type of ground, etc. To illustrate this point briefly, consider for
example Spanish, English and (Bokmäl) Norwegian. Spanish is often insensitive to
any particularity of the Locative scenario, whereas English and Norwegian are
more sensitive. The examples illustrate this phenomenon:
(39)
a. el libro EST-ά en hi repisa
the:Μ book be:located-3SG:PRES LOC the:F shelf
b. bok-en STÄ-r pä skjelv-et
book-the:C stand-PRES on shelf-the:N
c. the book IS on the shelf
These three examples predicate the static location of a referent, the book, with
respect to a ground, the shelf. In the Spanish example in (39.a), we encounter
generic information; both the verb estar and the preposition en render a generic
location. There is no further information as to what type of configuration the book
describes with respect to the ground. In English, the choice of the preposition on
construes the shelf as a surface on which the book is located. The verb to be here is
a general copula, less specific for this purpose than the Spanish verb. In Nor-
wegian, in contrast, the verb informs about the vertical posture of the book, a
nuance that lies unspecified in both Spanish and Norwegian. As English, Nor-
wegian construes the shelf as a surface.
As with the Ablative, markers such as these ones, regardless of their semantic
specification, are treated as Locatives. None of these languages can illustrate any
possible combination with an Agent. For this purpose, consider examples (40) and
(41) which exemplify the pattern <LOC-PASS-AG> in Evenki (AL), (NEDJAL-
KOV 1997), and (42) and (43) which illust :.'e the pattern <LOC-ERG> in Wagaya
(AT), (BREEN 1976c):
(40) The Sf. -du is the most common marker to express any Locative
construal, "regardless the size or orientation of the location" (p. 169):
turu-du "in Tura"; lamu-du "in the sea"; tatkit-tu "at/in the school";
ure-l-du "in the mountains"; oron-du "on a reindeer"; chuka-du "on
the grass"; etc.
(41) homo. ty bejumimni-du va:v-cha (bi-si-n)
bear hunter-D(/:PASS-AG kill-PASS-PRT be-PRES-3SG
'The bear is/was killed by the hunter' (p. 152)

55. Random documents with unrelated
content Scribd suggests to you:

56. So war er einige Stunden mit großer Anstrengung dahin geeilt,
als ihn seine Kräfte verließen und er an einer zufällig gefundenen
Quelle niedersank, die ihm zur Reinigung seiner Wunden, zur Stillung
des brennenden Durstes sehr wohlthätig wurde. Die Ermattung
machte ihn bald einschlafen.
[35] Siehe die Abbildung.

57. XXXV.
Die Ermordung eines
Handwerksburschen.
Laßt uns nicht stille stehen, denn geschäftig sind
Die Feinde rings, den Weg uns zu verschließen.
Schiller.
Sein Erwachen war nicht das angenehmste. Mariane, Sarberg,
Eckold, Schöneck und Schickel hatte er unter den Schüssen und
Hieben seiner Feinde niedersinken gesehen. Mit Recht mußte er
befürchten, daß auch Lehmann, Hentzschel und seine übrigen
Vertrauten die blutigen Opfer ihrer wüthenden Feinde geworden.
Auf eine Verbindung mit der Bande durfte er nicht mehr hoffen,
nur zu gut kennend den Haß des Welschen und den Einfluß, welchen
dieser tückische, gewandte Wohlredner auf die Bande hatte; ja Lips
Tullian war überzeugt, selbst von seinen frühern Kameraden, denen
er einst das Höchste ihrer Liebe, ihrer Treue, ihres Vertrauens
gewesen, dem nächsten Gerichte verrathen oder überliefert zu
werden, sobald er wieder zu ihnen zurückkehre.
Nach langer, ernster Selbstberathung bestimmte er sich dahin,
seine noch übrigen, in der Felsenschlucht vergrabenen Kostbarkeiten
zu sich zu nehmen, nach Wien zu gehen, und dort theils von den
Mitteln, die ihm der Verkauf dieser Kleinodien gewähre, theils vom
falschen Spiele so lange zu leben, bis sich ihm eine Gelegenheit

58. biete, eine kleine, tüchtige Rotte zu sammeln und an ihrer Spitze den
Pfad seiner bisherigen Industrie wieder zu betreten.
Für den Augenblick war ihm eine Kopfbedeckung und ein anderer
Anzug das Nöthigste.
Das Glück begünstigte den Bösewicht. Lips Tullian hatte sich
kaum von der Quelle entfernt, als er auf einen Fußpfad kam und von
ferne einen Handwerksburschen mit hoch aufgepacktem Reisebündel
daher kommen sah.
Der Handwerksbursche verrieth durch seine große, kräftige
Gestalt eine Körperstärke, mit welcher Lips Tullian, vom Blutverluste
und dem langen, eiligen Marsche geschwächt, die seinige nicht
messen zu können befürchtete.
Durch List mußte das Ziel erreicht werden. Er sah sich bisher von
dem Reisenden noch nicht bemerkt. Schnell kauerte er sich unter
einem Baume nieder, und begann zu stöhnen und zu wimmern.
Der Handwerksbursche wurde aufmerksam und schritt rascher
heran. Als er einen Verwundeten sah, der wie ohnmächtig vor
Schwäche und Schmerzen ächzte, fragte der Gutherzige mit recht
menschenfreundlicher Theilnahme, ob er einige Hülfe leisten könne.
Lips Tullian bat mit leiser Stimme und den Geberden des
Schmerzens und der Entkräftung, ihn nur einige Schritte tiefer in den
Wald zu führen, wo er eine Quelle wisse.
Der Handwerksbursche warf seinen dicken, schweren
Stachelstock nieder, und reichte Lips Tullian beide Hände, um ihm
empor zu helfen. Lips Tullian that, als könne er ohne Hülfe nicht
einen Schritt fortwanken. Er ließ sich den Stock geben, stützte sich
auf den Handwerksburschen und wankte so mit schleppenden Füßen
einige Schritte fort. Jetzt stand er, zog seine Uhr und ließ sie, als
hätten seine zitternden Hände nicht mehr die Kraft, sie einzustecken,
zur Erde gleiten.
Dienstfertig bückte sich der Handwerksbursche. Im Augenblicke
hatte Lips Tullian den wuchtvollen Stachelstock geschwungen, und

59. dem Arglosen solch’ einen mörderischen Schlag auf das Hinterhaupt
gegeben, daß er entseelt aufs Angesicht hinstürzte.
Schnell riß Lips Tullian den Mantelsack an sich, fand darin einen
fast neuen, vollständigen Anzug, feine Wäsche und manches gute
Kleidungsstück. In wenigen Augenblicken war er umgekleidet, und
der ganze Anzug paßte ihm vollkommen.
Nun durchsuchte er die Tasche des Todten, fand eine silberne
Uhr, einen Beutel mit Geld und eine Kundschaft. Er zog den
Ermordeten nackt aus, packte dessen Kleider in den Mantelsack,
seine eigenen aber warf er ins nächste Dickicht, hing ihn auf den
Rücken und zerschnitt und verstümmelte das Gesicht des Todten auf
das Gräßlichste, daß man den Unglücklichen nicht wieder erkenne,
vielleicht gar für ihn, Lips Tullian, zu halten veranlaßt werde, da er
sich gleich vorgenommen hatte, das Gerücht von seiner Ermordung
überall zu verbreiten, um auf seinen Wegen vor der Aufmerksamkeit
der Behörden und ihrer Diener mehr gesichert zu sein.
Ungeachtet der Einbruch der Nacht nicht fern war, so eilte er
doch nicht auf dem Fußwege fort, der ihn bald würde zu einem
Dorfe geführt haben, sondern schlug eine andere Richtung ein, um
Orte zu vermeiden, durch welche der Handwerksbursche gegangen
sein mochte. —

60. XXXVI.
Die Wiedererkennung.
Wie viele Kämpfe mußten wir bestehn,
Von wie viel Noth und Herzensangst ermatten,
Wie viele Leichname hinopfern und bestatten,
Eh’ wir uns hier in dieser Hütte sehn!
Schiller.
Schon oft hatte Lips Tullian in diesem Walde theils einzeln, theils
mit einigen der Bande Nächte zugebracht, aber immer nur im
Durchzuge, und wußte von der Lage und dem Umfange dieses
Waldes nur so viel, daß er in gerader Richtung nach Osten beinahe
zwei Tagereisen weit sich ausdehne. In dieser Richtung ging er nun
fort, mit ziemlicher Eile, da das volle Mondlicht für seinen Marsch
sehr günstig war.
Mit Anbruch des Tages stieß er auf eine Umzäunung, und eine
armselige, dunkle Strohhütte, daneben ein Meiler, sagten ihm, daß
hier ein Köhler hause.
Die Nähe eines Köhlers war ihm nicht unangenehm, da Leute
dieses Standes entweder größtentheils die Vertrauten der
Gaunerbanden sind, oder den Erzählungen und Angaben fremder
Leute solch’ einen vollen Glauben schenken, daß man von
Menschen, welche alles für blanke Wahrheit nehmen, allgemein
sagt: sie haben einen Köhlerglauben.
Ueberdieß glaubte er in der armseligen Hütte eine
Engelswohnung zu erblicken, da er vor Hunger, Durst und Ermattung

61. sich kaum mehr fortschleppen konnte.
Unentschlossen, ob er den Schlaf der Hüttenbewohner stören,
oder deren Erwachen hier abwarten solle, lehnte er sich an die
Umzäunung und musterte aus Neugierde und zum Zeitvertreib die
Kundschaft des armen Handwerksburschen.
Hatte er in seiner Raub- und Mordbegierde übersehen, daß der
Handwerksbursche wirklich viele Aehnlichkeit mit ihm habe, so fand
er diese in der hier aufgeführten Personalbeschreibung. Die Größe
der Gestalt, die Farbe der Haare, der Augen, die Gesichtsbildung
waren so bezeichnet, als hätte er selbst dem Kundschaftsschreiber
zum Muster gesessen; auch das Lebensalter traf auf ein Paar Jahre
zusammen, und Christoph Feller mußte gerade ein Schlossergeselle
sein, auf daß Lips Tullian mit dem Wenigen, was er von diesem
Handwerke wußte, doch in großer Sicherheit seine Wanderung
antreten konnte. Er hätte die Kundschaft nicht um eine hohe Summe
hingegeben.
In der Köhlerhütte wurde es rührig. Die Thüre that sich auf, und
eine hohe, schlanke Mädchengestalt im allerfreiesten Anzuge, eilte
mit leichtem, zierlichem Gange, ein Wassergefäß tragend, dem
Brunnen zu. Lips Tullian machte sich durch ein kleines Geräusch
bemerkbar.
Das Mädchen hielt an, betrachtete ihn aufmerksam, ließ den
Eimer fallen, sprang, wie ein flüchtiges Reh, über die Verzäunung,
und lief mit offenen Armen auf Lips Tullian zu. Sie drückte ihn mit
dem Lächeln einer sehr großen Freude an ihren üppigen Busen, sie
gab ihm heiße Küsse und nannte den Ueberraschten unter den
zärtlichsten Liebkosungen ihren lieben, lieben Tullian.
Der Dirne entging dieses Erschrecken nicht; sie beruhigte ihn
aber schnell, als sie sich ihm nannte. Nun erinnerte sich Lips Tullian
an Fräulein Margarethe, so wurde sie bei der Bande genannt, weil
sie die außereheliche Tochter eines Adeligen war.
Kaum vierzehn Jahre alt, von der liederlichen Mutter zu allen
Lastern gebildet, war Margarethe schon die Zuhälterin eines

62. Mitgliedes der schwarzen Garde.
Lips Tullian, der schöne, kräftige Mann mit der stolzen,
ehrfurchtgebietenden Haltung, das Haupt einer mächtigen Bande,
war für Margarethe das Ideal der Erhabenheit, ein Wesen ihrer
tiefsten Verehrung, ihrer innigsten Bewunderung, immer mehr ihrer
heißesten Liebe. Wo sie sich ihm nahen konnte, trat sie ihm
entgegen, um den Mann ihres Herzens zu sehen, um von ihm einen
freundlichen Blick, ein Lächeln, ein gütiges Wort zu erhaschen.
Lips Tullian, nur reife, rüstige Gestalten für seine Sinnenlust
liebend, übersah das Kind. Aber Mariane, den Geliebten immer mit
Argusaugen bewachend, durchschaute die Wünsche und die
Bemühungen der jungen Dirne.
Von Eifersucht gefoltert, machte sie durch ein vertrautes Weib
aus der Bande Margarethens Buhlen auf die Neigung seiner Dirne
zum Hauptmann aufmerksam; sie wußte Eckold in ihr Vertrauen zu
ziehen und ihn dahin zu verleiten, daß Margarethe als eine geheime
Kundschafterin des Gerichts angeklagt und ihre heimliche Ermordung
beschlossen wurde.
Im Branntweinrausche bestätigte Lips Tullian das Todesurtheil,
und Margarethe wäre das Opfer der grausamsten Eifersucht
geworden, hätte sie nicht das Glück gehabt, das morsche
Gitterfenster des Kellers, worin sie verwahret wurde, zu
durchbrechen und unbemerkt zu entfliehen.
Der Köhler, bei welchem sie jetzt lebte, fand sie, vor Hunger,
Durst und Ermattung beinahe am Rande des Grabes, kaum mehr zu
flüstern fähig, in der Nähe seiner Hütte unter einem Baume. Er trug
sie unter sein Dach, gab ihr Nahrung und Pflege, und nahm die
Verlassene als Tochter und Wirthschafterin an, da sein Weib und
seine Kinder schon lange hinübergegangen waren. Aus einer
liederlichen, arbeitsscheuen, buhlsüchtigen Dirne wurde Margarethe
unter der väterlichen Leitung des gutherzigen, frommen Köhlers ein
arbeitsames, sittiges Mädchen, das einfache Stillleben, die Ruhe, die
Einsamkeit des Waldes immer mehr liebend und den kleinen

63. häuslichen Geschäften, der Pflege ihrer wenigen Blumen und dem
schuldlosen, süßen Umgange mit der Natur sich aus voller Seele
widmend.
So waren beinahe vier Jahre verflossen, aber die Erinnerung und
die sehnsüchtigen Wünsche, Lips Tullian, den Gegenstand ihrer
heißesten Liebe, wiederzusehen, noch immer nicht ganz ihrem
Herzen entschwunden. —
Da stand der Mann vor ihr, dessen Bild sie in der Tiefe ihrer
fühlenden Seele so lebendig und so treu bewahrt hatte.
Das Mädchen war so reizend, daß der sinnliche Lips Tullian kein
Verlangen nach Labung, keine Schwäche fühlte. Er tauchte die
begehrenden Augen in die reiche Fülle des fast unverhüllten Busens,
der üppigen Formen, des blühenden Gesichtes voll Liebreiz; seine
Küsse, seine Umarmungen wurden immer feuriger, und Margarethe,
trunken von der Wonne des Wiedersehens, wäre schon in diesem
Augenblick das willige Opfer des Aufgeregten geworden, hätte nicht
die rufende Stimme ihres Nährvaters sie ihrem Sinnentaumel
entrissen.
Hastig flüsterte ihr Lips Tullian die Weisung zu, ihn als einen
Blutsverwandten, der von ihrem Aufenthalte in dieser Köhlerhütte
gehört habe, und vor seiner Wanderung ins Ausland zum
Abschiedsbesuche gekommen sei, dem Alten bekannt zu machen.
Der schlichte, treuherzige Waldmann glaubte alles, was
Margarethe über ihren Vetter mit geläufiger Zunge ihm erzählte,
reichte Lips Tullian wie einem alten Bekannten, mit einem herzlichen
Gruße die kohlenfarbige Hand und forderte gleich mit dem
gutmüthigsten Ungestüme, daß der willkommene Vetter seiner guten
Pflegetochter einige Tage bei ihm weile und mit dem sich begnüge,
was Armuth und guter Wille bieten könne.

64. L. Oeser in Neusalza.
Die Wiedererkennung.
❏
GRÖSSERES BILD
Lips Tullian wäre gern Jahre lang in dieser ärmlichen Hütte
geblieben, die ihm Margarethens Nähe und die Hoffnung auf so
manche süße Stunde schon jetzt zu einem Eden machten.
Es war am Morgen des dritten Tages, als der Köhler, der schon
mit der Morgenröthe in den Wald gegangen war, verdrüßlich

65. zurückkam, und Margarethen gebot, ihm seinen Sack, in welchem er
jederzeit bei längerer Entfernung vom Hause Lebensmittel mit sich
trug, mit Brod, geräucherter Wurst und Branntwein zu versehen.
Zugleich erzählte er, auf dem Arbeitsplatze von den herrschaftlichen
Jägern aufgesucht, und zum Streifzuge gegen die Tullian’sche Bande
befehligt worden zu sein, mit dem Beisatze, daß längstens in einer
Stunde die noch aufgebotenen Köhler, wie auch die Jäger und
Gerichtsdiener sich hier zur Versammlung einfinden werden.
Erbleichend und zitternd blickte Margarethe auf Lips Tullian hin,
der aber Besonnenheit genug hatte, die thätige Wachsamkeit der
hochpreislichen Obrigkeit höchst lobenswürdig zu finden, zugleich
aber erklärte, seines Bleibens dürfe hier nicht länger sein, weil er
sonst befürchten müsse, bei dem Meister, der ihn zur Arbeit
verschrieben habe, zu spät einzutreffen.
Während des Packens seiner Habseligkeiten, wobei sich
Margarethe sehr geschäftig zur Hülfe anließ, flüsterte sie ihm zu, um
die Hütte zu gehen, und am Holzstalle ihrer zu harren.
Lips Tullian nahm Abschied und hatte alle Mühe, dem Alten, der
sich ihm zur Begleitung bis zur Martersäule, von wo aus der Weg
nach der Landstraße nicht mehr verloren werden könne, aufdringen
wollte, von diesem Vorhaben abzubringen.
Er ging an den bezeichneten Platz. Nach einigen Augenblicken
kam Margarethe, legte eine Leiter an und bedeutete Lips Tullian
durch Winke, das Bodenloch des Holzstalles zu erklettern, und sich
dort bis zu ihrer Wiederkunft sehr ruhig zu verhalten. Als er oben
war, verbarg sie schnell die Leiter und schlüpfte in die Hütte.
Lips Tullian hörte aus seinem Verstecke das Herankommen vieler
Menschen, hörte oft seinen Namen mit den heftigsten
Verwünschungen nennen und wurde sehr unruhig, als die Leute sich
unter einander erzählten: es sei im Umkreise einiger Stunden solch
ein Zusammenfluß von Militär, Jägern, Gerichtsdienern und Bauern,
daß die ganze Gegend von Streifzügen wimmle.

66. Es wurde ihm wieder besser zu Muhte, als eine kräftige Stimme
den Aufbruch gebot und bald hatte sich das Geräusch der
Dahinziehenden in der Ferne verloren.
Margarethe gab ihm ein Zeichen, herabzukommen. Er wartete
nicht das Anlegen der Leiter ab, sondern schwang sich behend von
dem Balken herab. „Du mußt fliehen, auf der Stelle fliehen,“ — sagte
Margarethe, zog ihn in die Hütte, riß einen Schrank auf und packte
Kleider und Wäsche in einen Bündel — „aber ich fliehe mit Dir. Willst
Du, daß ich bleibe, so tödte mich, denn ohne Dich wäre mein Leben
ein ununterbrochener Tag der Trauer, des Schmerzes, der höchsten
Sehnsucht nach Dir. Ich bettle, ich stehle, ich morde für Dich, meine
Seligkeit opfere ich Dir auf, aber ich muß in Deiner Nähe sein. Ich
führe Dich einen Weg, wo kein Späher Dich ersehen, kein Häscher
Dich fangen wird. Ich kenne den Weg in jene Gegend, wo Du, wie
Du mir gestern vertrautest, verborgene Schätze besitzest, sehr
genau, da ich gerade in jener Gegend schon zweimal mit meinem
Nährvater war, der dort eine kleine Erbschaft erhob. Fast immer
durch Wälder leite ich Dich. Da, wo man Dörfer und Weiler nicht
umgehen kann, darfst Du mit deiner Kundschaft ohne Besorgnisse
wandern; ich nehme, um kein Aufsehen zu erregen, andere Wege,
und wir einigen uns wieder an bestimmten Orten. So, das Wenige,
was ich besitze, ist nun in diesem Bündel, jetzt laß uns die Reise
antreten!“ —
Schweigend hatte Lips Tullian Margarethens ihm wohlgefällige
Rede gehört, es wäre ihm gar zu schwer geworden, sich von der
reizenden Dirne zu trennen, und ihre Schlauheit, ihren Muth und
Gewandheit recht gut erkennend, glaubte er überzeugt sein zu
dürfen, daß ihre Gesellschaft für ihn einst sehr vortheilhaft werden
könne.
Er wanderte fort, und fröhlich und tändelnd hüpfte die
leichtfertige Dirne, aus dieser stillen, frommen Hütte, und gedachte
nicht mehr der Wohlthaten ihres Nährvaters, der guten Vorsätze,
denen unter seinen Lehren sich ihr Herz geweihet hatte, und schied
ohne Thränen von dem Orte, wo ein großmüthiger, gottesfürchtiger

67. Greis drei Jahre hindurch ein Vaterherz für die Verlaßne gehabt
hatte.
Die Felsenschlucht in der Oberlausitz wurde erreicht, der
vergrabene Schatz gehoben, und Lips Tullian beschloß, nach
Böhmen zu gehen, Prag zu meiden und dann durch einen Theil von
Mähren die Richtung nach Wien zu nehmen.

68. XXXVII.
Lips Tullian und Margarethe in
Prag.
Er seine alten Plane aufgegeben?
Ich sag’ euch, daß er wachend, schlafend mit
Nichts anderm umgeht.
Schiller.
Was noch von den Kleinodien aus dem Raube in dem gräflich
Beuchling’schen Palaste zu Dresden Lips Tullian in der Felsenschlucht
gefunden hatte, war von großem Werthe, aber Lips Tullian würde es
gern tief unter dem Preis dahingegeben haben, hätte er gleich
baares Geld dafür bekommen.
Mit einem Reichthume in seinen Taschen, der die Wohlhabenheit
einer ganzen Familie hätte begründen können, war Lips Tullian jetzt
in der Lage, kaum für sich und Margarethe den nöthigsten
Lebensunterhalt beschaffen zu können.
Das wenige Gelt, welches er bei seiner Flucht gehabt hatte, die
unbedeutende Baarschaft des erschlagenen Handwerksburschen,
und der Erlös aus dem Verkauf der beiden Uhren war in der
kürzesten Zeit dahin, denn Lips Tullian aß und trank gern das Beste,
und nichts war ihm zu theuer, wonach ihm gelüstete.
Auch mußte er drei Tage in einem Dorfwirthshause bleiben, da
Margarethe im muthwilligen Springen über einen breiten Graben den
Fuß verstaucht hatte.

69. Als ein Reisender ohne Kutsche, ohne Gefolge irgend Jemand
eine Schnur orientalischer Perlen, einen Ring mit großen Brillanten
vom ersten Wasser, ein juwelenreiches Armband zum Kaufe
anzubieten, wäre höchst unklug gewesen, da wohl eine strenge
Frage würde gedrohet haben, wie ein wandernder Schlossergeselle
zum Besitze dieser höchst werthvollen Kleinodien gekommen sei.
So gern Lips Tullian Prag vermieden hätte, so mußte er doch
dahin gehen, wohl wissend, daß es dort Leute genug, besonders
unter den Juden, gebe, denen er ohne Furcht vor lästigen Fragen
seine Sachen anbieten könne.
Zur Bestreitung der Reisekosten wurden nun Kleidungsstücke von
Wirthshaus zu Wirthshaus verkauft; Margarethens Reisebündel und
sein Felleisen lieferte die Mittel zu manchem leckern Gerichte und
mancher guten Flasche Wein. Selbst das Felleisen mußte noch für
die Schwelgerei in der letzten Nachtherberge vor Prag hingegeben
werden.
Nicht von dem mindesten Gepäcke belästigt, gleichsam wie von
einem Spaziergange heimkehrend, schlichen Lips Tullian und
Margarethe durch die dunkle Thorwölbung der uralten slavischen
Stadt.
Das kleine, verwitterte, dunkle Wirthshaus zum blauen Fuchs, in
einer wenig belebten Gasse der Altstadt gelegen, nahm die
Reisenden auf.
Unter den vielen Winkelschenken in Prag, wo das Gesindel
Zuflucht findet, kannte Lips Tullian aus seinem frühern Aufenthalte
den blauen Fuchs als eine vollkommene Raubhöhle, wo in der
Zechstube jeden Abend einige recht wackere Bürger sich bei der
Flasche, beim mäßigen Kartenspiele und mit traulichem Geplauder
vergnügen, während im Hintergebäude ein Paar kellerartige Gewölbe
von liederlichen Hausvätern und feilen Dirnen, von verdächtigen
Reisenden und lichtscheuem Gesindel belebt sind.
Mit Freude erblickte Lips Tullian in dem freundlich begrüßenden
Wirthe den nämlichen, der bei seiner Anwesenheit zu Prag hier die

70. Wirthschaft getrieben hatte, und auch ihn erkannte der Wirth nach
einem scharfen Beschauen, sich wohl erinnernd der hübschen
Summe, die bei ihm der schmucke Schlossergeselle vergeudet hatte,
daher sich der Ankunft des fröhlichen Gastes hoch erfreuend.
Ein traulicher Handschlag verbürgte stillschweigend die
Erneuerung sonstiger Vertraulichkeit, und als der Wirth Lips Tullian
fragte, wo er sein Gepäcke habe, hätte ihm dieser mit Bias
antworten können: „Omnia mecum porto.“ Er verständigte den
Frager nur flüchtig, daß sein Koffer nachkomme, und erbat sich mit
einem bedeutenden Winke ein apartes Zimmer, in welches ihn nebst
Margarethen der Wirth schleunigst führte.
Das Zimmer war recht bequem eingerichtet, der Tisch gleich mit
Wein und kalter Küche besetzt, und Lips Tullian nahm sich vor, hier
einige Tage nur der Tafel und Ruhe sich zu ergeben, und dann erst
den Verkauf seiner Kostbarkeiten zu besorgen, wie auch ein kleines
Nebengeschäft zu versuchen, um Prag nicht ohne Gewinn zu
verlassen. —
Am andern Morgen erzählte ihm der Wirth, daß er außerhalb der
Stadt noch einen kleinen Gasthof besitze, wo sich Abends junge,
reiche Leute versammelten, um in dem abgelegenen Gartenhause
Hazardspiele zu treiben, da diese in der Stadt seit einiger Zeit sehr
verpönt seien; übrigens werde dort Niemand zugelassen, für den
nicht der Wirth selbst, in Beziehung auf Verschwiegenheit und
Spielmittel, Bürgschaft leiste.
Lips Tullian, durch Verkauf seiner Kleinodien im Besitze einer sehr
bedeutenden Summe, und von der Lust gereizt, wie in Spaa
glänzend zu erscheinen, stattete sich und Margarethe mit Kleidern
und Pretiosen aus, in welchen nur Reiche und Vornehme sich zeigen
können.
Als Baron Horn mit Gemahlin, im Mecklenburgischen begütert,
wurde er der Gesellschaft von dem Wirthe vorgestellt. In den vier
Wochen seines Aufenthalts zu Prag besuchte er jeden Abend das
Gartenhaus, im Wechsel des Gewinnes und des Verlustes, je

71. nachdem er einen noch ärgern Betrüger als seinen Meister fand,
oder ein kluges Benehmen gegen die argwöhnischen Mitspieler ihm
Gewinn oder Verlust vorschrieb. Mit einer gewonnenen Summe von
430 Dukaten verließ er Prag.

72. XXXVIII.
Margarethens Untreue.
Des Mädchens Flamme währet,
Bis Lunens Hochlicht zweimal wiederkehret;
Dann sucht sie neuen Zeitvertreib.
Seume.
Unter Lips Tullians Spielgefährten zeichnete sich ein junger
Mensch durch Schönheit, durch die heiterste Laune und ein fast
ununterbrochenes Spielglück vorzüglich aus.
Schon in der ersten Nacht wurde Lips Tullian auf ihn sehr
aufmerksam, da er recht gut bemerkt hatte, daß dieser gewandte
Betrüger durch seine feinen Künste das Glück zu fesseln wisse.
Gleich hatte Lips Tullian die Idee, sich näher an ihn anzuschließen,
ihn zur Reise nach Wien zu bewegen, und zur gemeinschaftlichen
Sache in den dortigen geheimen Spielhäusern.
Daß der junge Mensch sich Baron von Strahl aus Litthauen
nenne, hörte er im Laufe der gesellschaftlichen Unterhaltung, von
dem Wirthe aber in vertrauter Mittheilung, daß dieser Baron Strahl
ein entlaufener Friseur aus Berlin, ein Matador unter den falschen
Spielern und vielseitig zu gebrauchen sei.
Lips Tullian wünschte sich Glück, seinen Mann gefunden zu
haben, und bald war unter beiden die innigste Freundschaft
geschlossen.

73. Nie besuchte Lips Tullian das Spielhaus, ohne von Margarethen
begleitet zu sein. In einer höchst freien Kleidung, mit süßem
Lächeln, mit feurigen Blicken und anmuthigen Bewegungen
wandelte sie an der Seite junger Gecken und alter Thoren, die nicht
des Spieles wegen hierher kamen, sondern mit den jungen
gefälligen Aufwärterinnen ihr Geld und ihre Gesundheit
verschleuderten, manche Stunde in den Gängen des Gartens auf und
nieder, fachte in den Gemüthern der Sinnlichen die heißesten
Begierden an, gab nie mehr als einen Kuß, eine Hoffnung für die
Zukunft, und wußte auch bei aller Härte der Versagung ihre Verehrer
so an sich zu fesseln, daß die Schlaue in kurzer Zeit sich im reichen
Besitze von werthvollen Ringen, Perlen und Stoffen sah.
Es war Lips Tullians Anweisung, die verschwenderische Großmuth
dieser Lüstlinge zu benutzen, und die gelehrige Schülerin übertraf
seine Erwartungen.
Aber dieses trügerische Spiel währte nicht lange. Margarethe
hatte für ihre freigebigen Verehrer bald kein süßes Lächeln, bald
keine bezaubernden Küsse, bald keine beglückenden Hinwinke auf
lohnreiche Zukunft mehr.
Ueber ihrer, bis zur Leidenschaft erwachsenen Liebe für den
schönen, immer fröhlichen Strahl vergaß sie die huldigende
Umgebung, ja, sie war von dieser Leidenschaft so beherrscht, daß
sie Lips Tullian offen gestand, ihm nichts mehr sein zu können, als
seine unerschütterliche Freundin, da ihr Herz, all ihr Sinnen, all ihre
Träume sich nur dem Geliebten weihen.
Lips Tullian, nur ein roher Wollüstling, der nach gestillten
Begierden auch die Reizendste mit Gleichmuth verließ, freute sich
dieser Mittheilung, da er mit Zuversicht hoffen durfte, durch die
Liebe und die Hingebung der reizenden Margarethe den jungen,
sinnlichen Menschen mit immer festern Banden an sich zu fesseln.
Strahl wurde auch von Margarethens blühender Schönheit, von
dem Zauber ihrer Anmuth, ihres lieblichen Frohsinnes und von dem
Feuer ihrer Umarmungen so begeistert, daß er jeder Gesellschaft,

74. jedem Genusse sich versagte, und nur für das Ideal seines Herzens
lebte.
Daß Lips Tullian ihn so viele Stunden mit Margarethen allein ließ,
daß er die Huldigungen nicht zu beachten schien, die der Liebende
der Geliebten immer sichtbarer weihte, hielt er für Kälte des
Ehemannes; er ahnte nicht in seinem Sinnenrausche, daß Lips
Tullians Benehmen auf seine Pläne sich gründete.
Mit ihm und Margarethen reiste Strahl von Prag ab, in einem
eleganten Wagen, den er um eine hohe Summe gekauft hatte, da
Margarethe einst den Wunsch äußerte, recht bequem und glänzend
die Reise nach Wien zu machen. Auch hatte der Gefällige, zu
Margarethens Bedienung, in Prag ein Mädchen gedungen, welches
die Reise nach Wien mitmachte und mit Strahl schon länger bekannt
zu sein schien.
Lips Tullian war der nachsichtigste Ehemann von der Welt.
Leckere Gerichte, gute Weine, ein trauliches Kosen mit Margarethens
Mädchen, das sehr schön und gegen ihn nichts weniger als spröde
war, nahmen den Schwelger zu sehr in Anspruch, um Margarethe
und Strahl zu stören.
Margarethe äußerte den Wunsch, über Olmütz und Brünn zu
reisen, um diese Städte zu sehen. Die Reise ging dahin. In Olmütz
kaufte Strahl ein Paar schöne flüchtige Wagenpferde, mit denen er
nun selbst fuhr, da, seiner Versicherung gegen Lips Tullian gemäß,
kein Kutscher sich gefunden hatte, der ihm tauglich geschienen
hätte. Es wurden nur kleine Tagereisen gemacht.
Lips Tullian sehnte sich nach Wien, und sah recht sauer, als Strahl
bei der Ankunft in Brünn erklärte, hier eine volle Woche bleiben zu
wollen, da er in der Nähe dieser Stadt einige Gutsbesitzer kenne,
denen er schon lange seinen Besuch zugesagt habe.
Strahl war auch wirklich während des achttägigen Aufenthaltes in
Brünn nur auf wenige Stunden sichtbar.
Endlich kam der Vorabend des Tages, mit dessen Anbruche die
Reise fortgesetzt werden sollte.

75. An diesem Abende wurde Lips Tullian von Strahl heimlich
gebeten, ihn, sobald Margarethe zu Bette sei, nach einem
Kaffeehause zu begleiten, wo man vortrefflichen Punsch trinke und
zwei allerliebste Mädchen nicht nur den Dienst der Hebe versehen,
sondern auch als Gnidias reizende Priesterinnen dem Opfer sich
weihen.
Solch eine Einladung von sich zu weisen, war Lips Tullian nicht
gewohnt. Er ging mit Strahl, und nachdem sie Arm in Arm mehrere
Hauptstraßen und eine Menge Seitengassen und Nebengäßchen
durchwandelt hatten, blieb endlich Strahl vor einer Spelunke stehen,
in welche Lips Tullian einzutreten zauderte, da er beim
Laternenschimmer umsonst nach dem bezeichnenden Mohren mit
der langen Gypspfeife und der Kaffeekanne oder nach einem
Aushängeschild sich umgesehen. Strahl hob seine Bedenklichkeiten
mit der Versicherung, dieser Eingang führe in das Hinterhaus, wo
man recht artige Zimmerchen, und darin Punsch und Mädchen, wie
auch recht angenehme, ungestörte Stunden finde, da der Billardsaal
und die Gemächer des Vorderhauses, des öffentlichen Anstandes
wegen, dergleichen Privatunterhaltungen nicht gestatten. Lips Tullian
folgte seinem Freunde, fand ein recht trauliches Stübchen, ein paar
schlanke, hochgeschürzte Nymphen in schamloser Kleidung und mit
frechem Entgegenkommen, fand vortrefflichen Punsch und eine
dieser Phrynen so anziehend, daß er unter ihren Liebkosungen
unmäßig trank und sich bald um seine Sinne getrunken hatte.

76. XXXIX.
Lips Tullian wieder an der Spitze
einer Räuberbande.
Laßt uns nicht stille stehen, denn geschäftig sind
Die Feinde rings, den Weg uns zu verschließen.
Schiller.
Lips Tullian erwachte aus todtähnlichem Schlafe. Er sah, er hörte,
er fühlte brennende Schmerzen am Kopfe, aber glaubte, noch immer
von den Irrlichtern eines neckenden Traumes in betäubender
Bewegung umtanzt zu werden. Nicht auf dem weichen Ruhebette
des kleinen traulichen Gemaches, nicht in den Armen einer
leichtfertigen Dirne fand er sich, sondern in einem spärlich
erleuchteten Gewölbe, auf einem Strohlager, einigen wilden
Gesichtern gegenüber, die sich auf dem Fußboden um eine Flasche
gelagert hatten und eifrig zusammen sprachen.
Alles trat ihm zu lebendig, zu wirklich entgegen; er raffte sich
zusammen, um den Armen dieses Traumgesichts sich mit Gewalt zu
entreißen, er riß sich auf von seinem Strohlager, und die furchtbare
Wirklichkeit empfing den Erstarrenden.
„Nun, Lips Tullian, hast Du endlich ausgeschlafen, damit wir ein
vernünftiges Wort mit Dir sprechen können?“ lachte eine dieser
Gestalten, und reichte ihm das volle Glas hin.
Lips Tullian war keiner Worte mächtig. Hier hörte er von einem
Unbekannten sich Tullian nennen. Wie war dieser Name, wie war er

77. selbst hierher gekommen?
Der Mann, von dem er so überraschend genannt wurde, mochte
wohl das Dunkel, in welchem Lips Tullian mit unsicherm Schritte sich
fortgriff, recht wohl durchschauet haben; er lagerte sich an des
sprachlos Staunenden Seite und sagte also:
„Eine lange, lange Erzählung könnte ich Dir zum Besten geben,
wie es so kommen mußte, daß Du hier bist; ich will Dich aber mit
wenig Worten klug machen. Dein Freund, Baron Strahl — sonst hatte
er einen andern Namen — und Deine treue Margarethe waren schon
in Prag darüber Eines Sinnes, Dich von der Last Deines Goldes zu
entheben, und mit ihrer Gegenwart Dich nicht länger zu
incommodiren.
Strahl schlug vor, Dich auf der Reise betrunken zu machen, in der
Kutsche zu erwürgen und Deine Leiche im nächsten Walde zu
vergraben. Darum kaufte er zu Olmütz eigene Pferde und nahm
keinen Kutscher an, damit Deine Ermordung um so unentdeckter
geschehen könne. Margarethe ging in diesen Vorschlag nicht ein,
und als Strahl darauf bestand, drohete sie, Dich zu warnen.
Jetzt besann sich Strahl, in dieser Gegend vor zwei Jahren
angegriffen, und dann selbst für einige Zeit einer unserer Cameraden
geworden zu sein. Er streifte in Brünns Umgebungen umher, fand
und erkannte mich, vertraute mir sein Verlangen, Dich aus dem
Wege geräumt zu wissen, und sagte mir zugleich, von Deiner
Zuhälterin erfahren zu haben, daß Du der berühmte Lips Tullian
seist. Ich erschrack vor Freude über diesen Namen, über Deine
Nähe, über Deine Bekanntschaft. Ich würde Strahl ermordet haben,
hätte er Dir Leides thun wollen.
Du sollst unser Bonherr werden, darüber war ich mit mir und
meinen Cameraden gleich im Reinen, und als Strahl versicherte, Dein
Reichthum sei zu bedeutend, um Dich den Aufenthalt in den
Wäldern und in schlechten Kneipen, die Gefahren, die
Anstrengungen unseres Handwerks für ein Leben voll
Bequemlichkeit, Ueberfluß und Genüssen eintauschen zu lassen, so

78. mußtest Du arm gemacht werden, um an unserer Spitze durch
Deinen Muth und Deine Talente, durch unsere Treue, Anhänglichkeit
und unsern Eifer wieder reich zu werden.
Die Sache war schnell gemacht. Strahl, durch mich unterrichtet,
führte Dich in dieses Häuschen, dessen Besitzer mein Bruder ist.
Trinken und Lieben, Deine schwache Seite, gaben Dich in unsere
Hände.
Strahl, Margarethe und ihr Mädchen, schon seit Jahren Strahls
geheime Zuhälterin, sind fort, Gott weiß, wohin, und mit ihnen Deine
Habseligkeiten und Dein Gold.
Du hast die Wahl, unser Anführer zu werden, oder in diesem
unterirdischen Gewölbe zu verschmachten!“ —
Die Wahl war nicht schwierig. Schon in der nächsten Nacht zog
Lips Tullian mit den neuen Gesellen hinaus in die Wälder, in das
wilde, blutige Räuberleben.

79. XXXX.
Lips Tullians abermalige
Gefangenschaft.
Nur zu! Sie rücken
Mit Schwert und Feuer auf uns an.
Schiller.
In einer Waldschlucht, zwei Meilen von Iglau, lag Lips Tullian mit
seinen Gesellen um ein hochaufschlagendes Feuer herum, an
welchem gesotten und gebraten wurde. Da gab es Geflügel aller Art,
einige Schafe, ein paar Kälber, auch an Wildpret gebrach es nicht,
und ein Fäßchen Branntwein versprach, das leckere Mahl zu würzen.
Aber die Räuber hatten nicht allein für den Magen, sondern auch
sonst für das Leben gesorgt. Einige ausgeraubte Kirchen, Edelsitze
und Bauerhäuser hatten viele gute und werthvolle Sachen in diese
Waldschlucht geliefert, und ein Säckchen mit Gold- und
Silbermünzen erfreute die glückliche Bande.
Die Kochtöpfe wurden in den Kreis der Hungernden gebracht, die
Braten vom Spieße genommen, und die Becher gefüllt. Gesang und
Geplauder verstummte, und nur das Löffelgeklapper und
Messergeklirre der Essenden unterbrach die Stille.
Da erscholl hinter Lips Tullians Rücken ein lautes Lachen, und als
er jetzt zurück blickte, meinend, einer der Kameraden wolle ein tolles
Stückchen erzählen, und durch eigenes Gelächter die Versammlung

80. 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