Download ebooks file Mediating between Concepts and Grammar Holden Härtl (Editor) all chapters

Visit https://ebookultra.com to download the full version and
explore more ebooks
Mediating between Concepts and Grammar Holden
Härtl (Editor)
_____ Click the link below to download _____
https://ebookultra.com/download/mediating-between-
concepts-and-grammar-holden-hartl-editor/
Explore and download more ebooks at ebookultra.com

Here are some suggested products you might be interested in.
Click the link to download
Revising Green Infrastructure Concepts Between Nature and
Design 1st Edition Daniel Czechowski (Editor)
https://ebookultra.com/download/revising-green-infrastructure-
concepts-between-nature-and-design-1st-edition-daniel-czechowski-
editor/
Intelligent Assistant Systems Concepts Techniques and
Technologies Roland H. Kaschek
https://ebookultra.com/download/intelligent-assistant-systems-
concepts-techniques-and-technologies-roland-h-kaschek/
Structural design for the stage Alys E Holden and three
others Second Edition (Online-Ausg.) Holden
https://ebookultra.com/download/structural-design-for-the-stage-alys-
e-holden-and-three-others-second-edition-online-ausg-holden/
Concepts in Toxicology 1st Edition John H. Duffus
https://ebookultra.com/download/concepts-in-toxicology-1st-edition-
john-h-duffus/

Starting Out The c3 Sicilian 1st Edition John Emms
https://ebookultra.com/download/starting-out-the-c3-sicilian-1st-
edition-john-emms/
Thomas Killigrew Cavalier Dramatist 1612 83 Alfred Harbage
https://ebookultra.com/download/thomas-killigrew-cavalier-
dramatist-1612-83-alfred-harbage/
Mediating Memory First Published 2018. Edition Bunty
Avieson
https://ebookultra.com/download/mediating-memory-first-
published-2018-edition-bunty-avieson/
Biomarkers of Environmentally Associated Disease
Technologies Concepts and Perspectives 1st Edition Samuel
H. Wilson
https://ebookultra.com/download/biomarkers-of-environmentally-
associated-disease-technologies-concepts-and-perspectives-1st-edition-
samuel-h-wilson/
Advances in Cancer Research Vol 83 1st Edition George
Klein (Editor)
https://ebookultra.com/download/advances-in-cancer-research-
vol-83-1st-edition-george-klein-editor/

Mediating between Concepts and Grammar Holden
HÃ¤rtl (Editor) Digital Instant Download
Author(s): Holden HÃ¤rtl (editor); Heike Tappe (editor)
ISBN(s): 9783110919585, 3110919583
Edition: Reprint 2011
File Details: PDF, 14.60 MB
Year: 2003
Language: english

Mediating between Concepts and Grammar
W
DE
G

Trends in Linguistics
Studies and Monographs 152
Editors
Walter Bisang
(main editor for this volume)
Hans Henrich Hock
Werner Winter
Mouton de Gruyter
Berlin · New York

Mediating between
Concepts and Grammar
Edited by
Holden Härtl
Heike Tappe
Mouton de Gruyter
Berlin · New York

Mouton de Gruyter (formerly Mouton, The Hague)
is a Division of Walter de Gruyter GmbH & Co. KG, Berlin.
© Printed on acid-free paper which falls within the guidelines
of the ANSI to ensure permanence and durability.
ISBN 3-11-017902-4
Bibliographic information published by Die Deutsche Bibliothek
Die Deutsche Bibliothek lists this publication in the Deutsche Nationalbibliografie;
detailed bibliographic data is available in the Internet at <http://dnb.ddb.de>.
© Copyright 2003 by Walter de Gruyter GmbH & Co. KG, D-10785 Berlin
All rights reserved, including those of translation into foreign languages. No part of this
book may be reproduced or transmitted in any form or by any means, electronic or mechan-
ical, including photocopy, recording or any information storage and retrieval system, with-
out permission in writing from the publisher.
Cover design: Christopher Schneider, Berlin.
Printed in Germany.

Contents
Mediating between concepts and language -
Processing structures 1
Heike Tappe and Holden Härtl
Mediating between
non-linguistic and linguistic structures
Coordination of eye gaze and speech in sentence
production 39
Femke F. van der Meulen
Time patterns in visual reception and written phrase
production 65
Philip Cummins, Boris Gutbrod
and Rüdiger Weingarten
Animacy effects in language production: From mental
model to formulator 101
Kathy Y. van Nice and Rainer Dietrich
Incremental preverbal messages 119
Markus Guhe
Word order scrambling as a consequence of incremental
sentence production 141
Gerard Kempen and Karin Harbusch
The linearization of argument DPs and its semantic
reflection
Andreas Späth
165

vi Contents
Semantics as a gateway to language 197
Heike Wiese
Mediating between
event conceptualization and verbalization 223
Temporal relations between event concepts 225
Elke van der Meer, Reinhard Beyer, Herbert Hagendorf,
Dirk Strauch and Matthias Kolbe
Segmenting event sequences for speaking 255
Ralf Nüse
Events: Processing and neurological properties 277
Maria Mercedes Pinango
Aspectual (re-)interpretation: Structural representation
and processing 303
Johannes Dölling
Type coercion from a natural language generation point
of view 323
Markus Egg and Kristina Striegnitz
The mediating function of the lexicon 349
The thematic interpretation of plural nominalizations 351
Veronika Ehrich
Competing principles in the lexicon
Andrea Schalley
379

Concepts of motion and their linguistic encoding
LadinaB. Tschander
Too abstract for agents? The syntax and semantics of
agentivity in abstracts of English research articles
Heidrun Dorgeloh and Anja Wanner
Index of names
Index of subjects

Mediating between concepts and language -
Processing structures*
Heike Tappe and Holden Härtl
1. Modules and interfaces
One of the main functions of language is to abstract over complex
non-verbal message structures. The language system generates
highly compact linguistic material which, however, must still enable
the recipient of the corresponding linear grammatical sequence to
fully infer the intended message. To guarantee this a device is re-
quired which links concepts and grammar in a systematic fashion by
negotiating the requirements of both the generalized linguistic struc-
tures and the underlying conceptual complexes. Typically, this me-
diating function is instantiated by an interface. Any interface device
has to satisfy procedural requirements because linguistic structure
building must accommodate the fact that different types of informa-
tion are available at different points in time.
Regarding aspects of design, an interface is a virtual or an actual
surface forming a common boundary between independent func-
tional units. It can be defined as a point of information transition and
communication. In a technical sense, an interface definition encom-
passes rules for information transfer and calls for a characterization
of the kind of data that can be handed over from one unit to the
other. This also entails the specification of structure-sensitive opera-
tions over those representations that are the output structures of one
functional component and serve at the same time as input structures
for the subsequent component. The diction independent functional
unit is akin to the term module, in that both notions imply a more or
less autonomous and specialized computational system to solve a
very restricted class of problems and uses information - which are its
proprietary - to solve them (cf. Fodor 1998).

2 Heike Tappe and Holden Härtl
In cognitive science it is widely held, that at least some human
cognitive mechanisms are organized in modules.1
Fodor (1983) de-
fines them as cognitive systems characterized by nine criteria, some
of which concern module-internal information processing with im-
plications for how the interface between such modules is to be de-
fined. The most prominent of these criteria are informational
encapsulation and domain specificity, meaning that; first, the inner
workings of a module cannot be directly influenced from the outside.
Second, that each module computes information of one distinct type,
which, however, has to be of tremendous significance to the species.
Further characteristic features are the following: Unconsciousness,
i.e. module-internal processing is opaque to introspection. Speed and
shallow output, which characterize modules as extremely fast cogni-
tive sub-systems producing a particular output, albeit without provid-
ing information about the mediating stages preceding it.
Additionally, modules are processing pre-determined inputs, which
in turn result in pre-determined outputs devoid any contextual influ-
ence (obligatory firing)} Since it was advanced Fodor's notion of
modularity has stimulated a vivid controversy and an enormous body
of research. In particular the idea of information encapsulation has
become fundamental to computer science. Many standard technolo-
gies of programming are based on this feature. Modularity also plays
a key role in artificial intelligence and computational linguistics: To-
day even systems within sub-symbolic intelligence such as neural
network systems depart from their traditional homogenous architec-
tures and use somewhat modular approaches especially so to natural
language processing (cf. McShane and Zacharski 2001). While it is
thus largely agreed upon that the human mind/brain is organized into
domain specific components (except in rigorous connectionist ap-
proaches), it can be witnessed that the current interpretation of the
term module varies immensely depending on the underlying general
framework (cognitivist, neuro-psychological, evolutionary connec-
tionist, etc.). Generally, it seems that Fodor's modularity assump-
tions are only partly shared in existing models of the human
mind/brain, i.e. the proposed modules are not usually held to possess
all nine Fodorian criteria.

Mediating between concepts and language - Processing structures 3
The related question whether the human language system is
carved up into functional units and - more strongly - whether these
or some of these are full-fledged modules in Fodor's sense has been
a hotly debated question in linguistics, philosophy and psychology
over the last two decades. Because of space limitation we cannot
reiterate this intricate discussion (but cf. e.g. Karmiloff-Smith 1992,
Marshall 1984, Frazier 1987a, Smolensky 1988 and Müller 1996 for
varying viewpoints on modularity). Generally, this thematic complex
is closely connected with a persistent delimitation effort in linguis-
tics. It is broadly held indispensable for both the definition of the
discipline and for scientific distinctness to accomplish an analysis of
language as a formal system. This endeavor dating back to de Saus-
sure (1916) has had its reflex in syntactic and semantic theory alike.
Consequently, the predominant position subsumes under the term
syntax language specific competencies of how symbols of some lan-
guage may be combined independent of meaning, of other cognitive
computations, and of socio-cultural requirements (cf. e.g. Chomsky
1986). Likewise, formal semantics strives to explicitly identify those
aspects of meaning that are genuinely linguistic, i.e. abstract-able
from general world knowledge, and at the same time persistent in all
syntactic alternation contexts (cf. Cresswell 1978, Montague 1970,
for an overview Bäuerle 1985). In the consequence formal ap-
proaches in linguistics have to date been primarily engaged in con-
sistently explicating language internal structures.
Starting in the 70ies, research in cognitive science, anthropology,
and psychology inspired approaches that deny the autonomy of syn-
tax - and of linguistic subsystems in general - in relation to concep-
tual structure. They interpret grammatical phenomena in terms of
more general cognitive principles with applications outside lan-
guage. These have been subsumed under the terms of cognitive
grammar and functional grammar (e.g. Bates and MacWhinney
1984, Deane 1992, Lakoff 1991, Langacker 1987, 1990, 1991,
Gärdenfors 2000, and Tomasello 2000a, 2000b). Without the as-
sumption of functional units that are engaged in some kind of divi-
sion of labor, the notion of a restrictive mapping device becomes
superfluous as the different parts of the language faculty are concep-

tualized as being highly interactive and having access to basically
the same information and knowledge sources.
The epistemological question, whether a formalist or a functional-
ist conception is preferable, gains in relevance when we take into
consideration language processing. The overarching endeavor to de-
velop models for language production and comprehension systems
calls for a specification the relevant sub-components and carries in
itself the need to describe and to explain the interaction of informa-
tional sources. This objective is characteristic for approaches that
attempt to preserve some of theoretically sound and the empirically
founded assumptions of theoretical linguistics and to incorporate
them into a language processing framework (cf. e.g. Levelt 1989,
Bierwisch and Schreuder 1992). Language production and language
comprehension processes are based on representations, on which
they operate. The computation of the linguistic meaning and thus the
communication of information are impossible without an accessibil-
ity of both general and linguistic knowledge. From this follows the
prime question: Which kinds of information interact in what fashion
and at what points in time during language processing? What we are
addressing here is the processing criterion, i.e. are the representa-
tions a given linguistic theory proposes computable by a language
processing system (Marcus 1982, Fodor 1983, Frazier 1987b, Fra-
zier, Clifton and Randall 1983, Berwick and Weinberg 1983). This
means that if we assume that grammars are theories of abstract lin-
guistic competence (e.g., Chomsky 1986), we have to ask whether
they may or may not provide an appropriate framework for under-
standing the mental processing of language (Stillings et al. 1998:
435).
Unfortunately, the discussion between different schools in lan-
guage research remains - as Newmeyer (1998) points out - to date
largely unsatisfactory. They tend to avoid direct confrontation and
thus they generally are unaware of the compatibility of their results.
For the most part this observation also characterizes the interdisci-
plinary communication on matters of modularity and in the conse-
quence on the structures and processes, which play a role at the
interfaces in question. While the understanding of how the linguistic

and the non-linguistic system interact, constitutes one of the most
interesting and central questions in language research, both an intra-
disciplinary and interdisciplinary convergence seems to be a long
way off. The respective definitions of the interface between grammar
and concepts - as well as its allocated character and scope - vary
substantially subject to the vigorousness of the underlying modular-
ity assumption. In the well-established Levelt model (1989) - that
has provided the reference architecture for the majority of research
in language production - the most intensively discussed interface
representation is the so-called preverbal message. In the rigorous
interpretation it links non-linguistic and linguistic structures. How-
ever, the question whether the preverbal message itself is to be inter-
preted as purely non-linguistic is to date still hotly debated. And, in
the consequence there exist profound controversial assumptions
about its general character and content. From this follows that the
impact of features in the preverbal message on the subsequent repre-
sentations remains under discussion especially regarding the realiza-
tion of this information by the sub-components of the linguistic
system. This concerns e.g., the question whether the linguistic reali-
zation of a preverbal message such as the word order of the utterance
is determined by the order in which concepts are selected, or, is the
outcome of purely grammatical operations.
In order to enhance both intra- and interdisciplinary exchange
about these issues, the current volume brings together researchers
both from theoretical linguistics and from language processing as
well as researchers from adjacent disciplines such as computer-
science and psychology. While all contributors acknowledge some
division of labor between lexical(-semantic), morphological, syntac-
tic, and phonological structuring, it is not surprising that they do not
define the respective sub-components and their substance in the
same way. Especially the term semantics receives different interpre-
tations as notions relating to meaning have long and often controver-
sial histories within the disciplines that contribute to this volume,
which are related to foundational and methodological differences. As
a consequence, the current volume comprises contributions that a
traditional perspective on the interface function in question would

not integrate. Although findings from language comprehension stud-
ies are also discussed, the main body of contributions center around
aspects of language production. In this field available definitions of
the concrete interaction between the conceptual/semantic and the
grammatical level are to date still of a tentative nature. The dispute
in the book will shed light on this issue by exploring the several
stages of processing ranging from the conceptual knowledge, its re-
cruitment, and preverbal preparation for linguistic computation, to
finally its grammatical realization.
In the following paragraphs we give an overview of prominent -
and in the interest of space selected - interface conceptions from the
perspectives of both the grammatical and the conceptual systems and
relate those to questions of language processing. Subsequently, we
introduce the contributions to this volume, which demonstrate vari-
ous parallels and common attitudes in spite of differences in focus,
research background, and modeling.
1.1. Linking to syntax
The assumption that a linguistic capacity of the human mind/brain
enables speakers to competently master their native language is
tightly intertwined with the influential work of Fodor (1983) and
Chomsky (1986). Both assert the existence of a specialized language
faculty, which is conceived as a mental organ3
and as being internal-
ly organized into several functional subsystems. Especially Chom-
sky's arguments in favor for a linguistic module are based on pheno-
mena which are hard to explain on other but syntax-internal
grounds.4
Further compelling evidence for genuine linguistic syntac-
tic principles are found in language acquisition (e.g. Meisel 1990,
Stromswold 1992, Tappe 1999) and Creole language data (e.g. Bick-
erton 1990).5
The division of the cognitive system into functional
sub-components implies the existence of specific principles organiz-
ing the representations within each component. More importantly in
the present context, it follows from this conception that mapping
mechanisms between the components be specified.

It is generally acknowledged that for a successful coupling be-
tween (lexical) semantics and syntax predicates have to provide such
lexical information as the number of arguments and the syntactic
structure into which these arguments are to be integrated. In spite of
this broad consensus, the proposals about how such an interrelation
between syntactic and semantic structures may be realized vary sub-
stantially.
Recent syntactic theories characterize syntactic operations by mi-
nimalist principles, which are subject to directives of economy and
explicitness. In the minimalist framework (cf. e.g. Chomsky 1995)
lexical items enter the syntactic building process fully equipped with
their grammatically relevant features including categorial, semantic
argument structure, and thematic features. The relevant operation
select maps lexical items from a set of elements activated from the
lexicon onto the computational process. This process makes use of
two basic mechanisms, i.e. merge and move. Furthermore, procrasti-
nate regulates that syntactic movement has to take place as late as
possible in the derivation, if there is a choice, which differs from
language to language thus creating language-specific word order
variations. The underlying idea is that covert movement is 'less
costly', because it does not have to pied-pipe phonological features
(cf. e.g. Chomsky 1995, Wilder and Cavar 1994). In this fashion the
syntactic component produces structures that are compatible and
legible to the linguistic levels adjoining the syntactic level and also
to the levels adjacent to the linguistic system itself. The language
faculty has to meet specific interface conditions to allow for interac-
tion with the adjoining nonlinguistic components. This requirement
has led Chomsky (2000) to the conclusion that "language is an op-
tional solution to legibility conditions". These legibility conditions
have to involve principles of how syntactic material is to be mapped
onto phonological representations of the articulatory-perceptual sys-
tem on the one hand, and the semantic representations of the concep-
tual-intentional system on the other.
Developing a somewhat different approach to modeling the lexi-
con-syntax interface within the feature-checking framework of the
minimalist program (Chomsky 1995), van Hout (1996) proposes a

CHecking Event-Semantic Structure model (CHESS). She assumes
that the event structure of a predicate must be syntactically identified
(cf. Grimshaw 1990; Grimshaw and Vikner 1993) and defines the
mapping relation in terms of checking event-semantic features in
functional configurations. There are two structural argument posi-
tions: the specifier positions of AgrS and AgrO. An argument in ei-
ther of these positions identifies an event or subevent by referring to
an event participant that is involved in that (sub)event. Telic event
type features must be checked in AgrOP. Van Hout argues that the
CHESS model accounts for the event-semantic mapping generaliza-
tions in a natural way, explaining the phenomenon of lexical-
syntactic flexibility as a derivative of event-type shifting.
These current developments within syntactic theory are compati-
ble with semantically oriented approaches that assume specific link-
ing mechanisms operational between semantic and syntactic
structure. Here it is held that specific configurational constellations
in the semantic representation determine the syntactic realization of a
language. In Bierwisch (1986) and Wunderlich (1997) the mapping
of arguments onto syntactic structure is organized through the em-
bedding of the arguments in the semantic form representation, i.e. a
predicate-argument structure. Jackendoff (1990) advances a similar
approach with the difference that he assumes correspondence rules
to negotiate between syntactic and semantic-conceptual structure.
Moreover, he also claims that lexical syntactic representation of a
predicate can always be reduced to its lexical semantic representa-
tion. In the consequence he treats the semantic and syntactic infor-
mation of the lexicon as part of conceptual structure whereby
arguments correspond to ontological categories of conceptual struc-
ture.
This latter claim differentiates Jackendoff s account considerably
from most linking theories. Based on the observation that some pairs
of predicates like, e.g. ask and inquire have different syntactic subca-
tegorizations albeit their semantics are identical, Grimshaw (1979)
proposes that predicates select both syntactic objects (nouns phrases,
sentences and semantic objects (propositions, questions, exclama-
tions) with no correlation between the two. The linking between the

two distinct types of information is handled by thematic hierarchies
where semantic argument features like AGENT, BENEFACTIVE or
THEME organize the order of arguments to be realized in syntax (cf.
Baker 1997, Grimshaw 1990, Jackendoff 1972 among many others).
AGENTS, for example, surface in a hierarchically higher position (as
subject) than THEMES (as direct object in transitive verb complexes).
The very nature of argument structure is less than clear.6
'Linking
theoreticians' assume that argument structure not only contains the-
matic information but that it is also closely tied up with event struc-
ture, which contains aspectual information (cf. Grimshaw 1990).7
Tenny (1992, 1994) assumes that only aspectually relevant informa-
tion is mapped onto syntax {Aspectual Interface Hypothesis). In the
other extreme, researchers like Rappaport and Levin (1988) encode
no more than syntactically relevant information into argument struc-
ture, which thus does not contain any thematic role specifications.
As becomes evident from this discussion, most of the various exem-
plary conceptions of the mapping between syntax and semantics are
joined by the consistent assumption that there is an independent
level, where lexical properties such as predicate-argument structures
are calculated. However, the question of what kind of information
influences and/or is to be integrated into this structure during lan-
guage processing has not yet received a widely accepted mutual an-
swer. This is partly due to the fact that syntactic theories tend to
center around the outcome of the computation rather than a real time
piecemeal construction of syntactic strings. In this context, the ques-
tion of how information is weighted such that the salience of the
constituents has its reflexes in an incremental syntactic realization
gains central importance.
1.2. Semantics
As was already hinted at in the first paragraph, formal model-
theoretic approaches towards meaning assume a modular organiza-
tion of linguistic processes: A morpho-syntactic component generat-
ing overt linguistic sequences and a semantic component, which re-
lates the grammatical material to extra-linguistic structures. General-

ly the focus of investigation the pairing of syntactic categories and
semantic types and the subsequent model-theoretical interpretation
of the analyses (e.g. in the framework of categorial grammar,
Ajdukiewicz 1935). The prime target is to specify how linguistic
expressions fit the world. Therefore investigations center, first,
around referring expressions, (syntactically encoded in noun
phrases) and, second, around truth-conditions of propositions, in-
cluding the exploration of which inferences follow from a linguistic
expression (cf. e.g. Lewis 1972, Tarsky 1977). Under this perspec-
tive the linking between syntax and semantics the need to further
explicate the linking between syntax and semantics does not arise
because here syntactic structures are considered categorical com-
plexes, whose interpretation is derived compositionally from either
the syntactic parts or their fixed meaning, respectively.
Syntactic constellations are deemed relevant only if the modifica-
tion of a linguistic string results in different entailments such that the
truth conditions underlying the expression in question are altered.
Correlations between certain linguistic expressions are taken to be of
a logical rather than a grammatical nature (cf. Montague 1973, Par-
tee 1975, Dowty 1979). Grammatically different but logically identi-
cal sentences inducing parallel entailments like the three examples in
(1) are generally treated in a homogenous fashion. The differences
between them are ascribed to information structure and focus pack-
aging routines.
(1) a. Somebody killed the fly.
b. The fly was killed.
c. The fly, somebody killed.
Somebody did something.
Decompositional approaches strive to grasp further entailments that
cannot be explicitly derived from overt form, but need to be inferred
from inherent meaning features. To this aim they employ the concept
of basic meaning components (cf. Katz and Fodor 1963, McCawley
1971 and many others). Under the assumption that complex mean-
ings are built up from smaller units such as CAUSE or NOT ALIVE,

more specific entailments can be logically derived from the sen-
tences in (1), cf. examples in (2).
(2) a. Somebody killed
thefly.
b. The fly was killed.
c. The fly, somebody
killed.
CAUSE[somebody,[BECOME[-iALiVE fly]]]
Somebody did something
Somebody caused something to
happen
Something became not alive
Although purely logically oriented, decompositional approaches can
thus capture implicit entailments, they cannot address the issue of
contextually driven truth evaluations. Under the assumption that - in
order to adequately convey a message structure - such information
structural values determining an expression are to be defined as re-
flexes of the speaker's intention, a broader notion of what is meant
by the term propositional content is needed. Consequently, the truth
conditions underlying the example in (2c) have to imply that this
sentence can have been uttered only in a specific contextual (i.e. a
contrastive) situation: The respective discourse set needs to contain
at least one more object such that the contrastive function of the ex-
pression can be evaluated as true.
A further shortcoming of purely logically oriented semantic theo-
ries is that they have to define truth conditions that must hold in
every possible situation the corresponding expression occurs in. For
example, a semantic analysis for short passives - cf. examples in (3)
- has to explain the fact that passives can be accompanied by pur-
pose clauses, which imply that there is an implicit agent denoted in
the matrix clause. This leads to the conclusion that the truth condi-
tions underlying passives have to signify an (existentially bound)
individual (cf. Brody and Manzini 1988, Roeper 1987, Koenig and
Mauner 1999, for discussion).
(3) a. The letter was written in order to impress the duchess.
b. The letter was written but it never reached its addressee.

In (3a) the implicit agent of the purpose clause (the one who im-
presses) and the demoted entity in the matrix clause (the one who
wrote the letter) are co-referential. Although this can surely be taken
as evidence for the conceptual existence of an implicit agent in short
passives, nothing prevents us from rejecting this assumption in cases
like (3b) where no purpose clause is added. However, the latter hy-
pothesis can merely be upheld if we assume a level of language
processing where only those pieces of information are provided
which are relevant for a successful realization of the communicative
act. Consequently, for a message like (3b) an implicit agent - as it
does not gain any referential salience - might not be present in the
semantic-conceptual structure underlying the message. Only in cases
where a conceptual activation of a corresponding entity becomes
relevant (as in (3a)) this knowledge has to be retrieved. Yet, in order
to cover cases where contextual constellations indeed require the
conceptualization of an entity, truth-theoretic analyses over-generate
and represent both sentences alike. Obviously, this problem concerns
the notion of conceptual activeness and here empirical and proce-
dural evidence may provide a solution by indicating the concrete
conceptual constellations holding during actual language processing
in real time. Against this background, it is apparent that experimental
results can not only help to reveal stages of language processing and
to define an adequate processing model, but also to indicate how
linguistic expressions be analyzed and to determine corresponding
representations.
Generally, semantic theories are, of course, not oblivious of the
importance of context- and situation-dependent aspects of meaning
construction, as is emphasized, e.g. in situation semantics (cf. Bar-
wise and Perry 1983, among others). Here, sentence meanings are
built up compositionally as functions from reference situations to
described situations. Thereby contextual factors reflecting specific
speech situations are incorporated into the study of meaning such
that expressions like I, this, and yesterday in I saw this plate on the
table yesterday are evaluated against the context of the actual speech
event. In this way, adequate means to determine corresponding truth-
values are provided. In a similar fashion, Kaplan (1977) distin-

guishes between fixed context-independent character of an expres-
sion and its content evaluation. The latter concept accounts for the
fact that the meaning of linguistic units is adapted to contextual re-
quirements and acknowledges that the interpretation of indexical
expressions like (demonstrative) pronouns is dependent on time.
In contrast to the model-theoretic approaches sketched above,
semantic theories that include grammatical aspects into the analysis
of linguistic expressions are enabled to explain entailment relations
between sentences that are based on lexical and morpho-syntactic
constellations. Consider the following examples:
(4) a. John broke the mirror.
b. John destroyed the mirror.
The mirror broke.
'The mirror destroyed.
The difference between destroy and break can be put down to inher-
ο
ent features of the respective lexical entries. Levin and Rappaport
Hovav (1995), for example, argue that only those verbs detransitiv-
ize which can express a change of state coming about without the
intervention of a volitional agent, i.e. which can instead denote an
effect of a natural force. In this sense, lexical semantics seeks to de-
fine predictable relations between semantic features and overt
grammatical behavior, which, at the same time, allows to predict
possible semantic relations between sentences such that a transitive
verbal complex entails the corresponding intransitive one and vice
versa.
Likewise, decompositional lexico-semantic approaches control
the mapping of grammatically relevant aspects of meaning structures
onto linguistic form by encoding grammatically visible differences
in meaning by way of decompositonal representations, which are
linked to morpho-syntactic representations. As we pointed out in
paragraph 1.1, for now there is still no agreement on the question
whether meaning aspects visible in grammar are to be defined as a
subset of the conceptual, non-linguistic level of language processing
or rather as part of the linguistic system. The former assumption im-
plies that conceptual structures are directly linked to syntactic struc-
ture - a view that is employed by conceptual semanticists like

Jackendoff (cf. Jackendoff 1992, 1997). Here, conceptual structures
that constitute the non-linguistic message have to be compatible with
both the linguistic system with its independent language-specific
requirements on the one hand, and the conceptual knowledge base
organizing information from the several sensory and memory sys-
tems on the other hand. In contrast, a more modular conception of
the encoding of grammatically relevant aspects of meaning is incor-
porated in theories that assume a separate, lexico-semantic level,
which is organized by strictly linguistic principles. This grammati-
cally determined level - the semantic form - of meaning representa-
tion is distinguished from a non-verbal, conceptual level comprising
propositional information of a message level by semanticists like
Bierwisch (1983), Dölling (1998), Ehrich (1992), Härtl (2001), Lang
(1994), Olsen (1998), and Wunderlich (1997). Similar distinctions
have been formulated in Mohanan and Wee (1999), who differenti-
ate a semantic structure from a conceptual structure, or Grimshaw
(1993) who distinguishes between the semantic content of an expres-
sion and its semantic structure. Similarly, the logical form level (LF)
of syntactically reflected meaning aspects such as the scope of quan-
tifiers or of negations in the Government & Binding program and its
successors (Chomsky 1981, 1993 and many others) can be consid-
ered a reflex of the need for a linguistically determined level of se-
mantic information. These rules generate semantically adapted
structures, which then are mapped onto representations of the con-
ceptual-intentional system of the conceptual knowledge base inter-
facing the several conceptual subsystems that organize the world-
knowledge of an individual.9
While these conceptions are in them-
selves quite elaborated, they are still largely oriented towards the
linguistic representations as outcome of processing stages, while the
processing aspects themselves are largely ignored. In language pro-
duction research, however, it is of prime importance to clarify how
conceptual structures might influence the construction of linguistic
representations and thus also the variability of semantic and syntactic
structures.

1.3. Conceptualization
As has become apparent in paragraph 1.2 cognitively oriented se-
mantic theories are primarily concerned with the question of how
semantic representations systematically interface non-linguistic and
syntactic representations. More broadly considered this is a common
goal in the interdisciplinary research aiming at understanding the
language faculty and its interaction with other cognitive capacities.
In this context the basic ontological categories, i.e. objects and
events, and how their respective conceptualization relates to verbali-
zation and comprehension are of prime importance. Growing evi-
dence from psychological and neurological research indicates that
objects and events cannot only be differentiated on philosophical and
theoretical grounds, but that the neural processing of these two basic
entity types engages discriminable sub-parts of semantic memory
(cf. e.g. Caramazza 1997).
Being able to talk about an object or to decode a specific object
reference has as its prerequisite object recognition. This complex
mental operation involves two more basic processes concerning ob-
ject constancy and object categorization. The first one relates to sta-
bility of object recognition independent of spatial transformations,
i.e. regardless of a given object's orientation, size and position.10
The
second one-object categorization- involves the ability to perceive
and categorize different objects as members of the same category. In
order to be able to tackle the second task, perceptual or conceptual
equivalences among the objects within a given class have to be de-
tected (cf. e.g. Anderson 1991, Bloom 1998, Medin 1989).
For the contributions to the current volume these two cognitive
processes are less important than the fact that humans generally ex-
perience objects in various locations and in many different spatial
arrangements. Consequently the spatial configurations in which ob-
jects occur and the spatial relations that hold between different ob-
ject become essential for linguistic encoding of situations. In
verbalization and in comprehension spatial relations between ob-
jects, which may freely employ the multidimensionality of space,
have to be linked to a linear string of linguistic expressions. Verbal

expressions typically contain projective expressions (e.g. left, right)
that are dependent on a specific perspective reflecting a view point
on the described situation. Perspectives are linguistically encoded by
utilizing reference systems, i.e. systematically structured fields of
linguistic expressions. Spatial reference systems are usually subdi-
vided into two major classes. Egocentric reference systems are those
in which relations between objects are specified in relation to body
coordinates of an observer (most prominently body-axes or retinal
coordinates). In environmental reference systems, on the other hand,
locations are characterized via objects other than the speaker; exam-
ples are absolute reference frames employing cardinal directions
{North, South, East, West), or, reference systems making use of
prominent landmarks (e.g. 'hillwards') (cf. Levinson 1996). As has
been pointed out in the literature, the employment of spatial perspec-
tives on a given situation is influenced by various parameters and
often is not maintained throughout a description (cf. e.g. Taylor and
Tversky 1996, Tappe 2000).
Object conceptualization also plays an essential role in event con-
ceptualizations, as in events entities figure as event participants.
Fundamental features of event structure must be accessed to assure
language processing, which e.g. determine during comprehension
which syntactic structure is projected. Depending on whether or not
the speaker/hearer identifies an initiator of the event, the verb class
will vary. A verb like, e.g. push, requires an initiator (which means
at the same time that it is always transitive), whereas break may or
may not encode an event with an initiator (i.e. may also be intransi-
tive). Another feature concerns whether there is an endpoint of the
event (telicity). Telic events must have an underlying direct object
(cf. O'Brian, Folli, Harley and Bever, in prep.).
In the larger context of event conceptualization the influence of
conceptual features like-most prominently-animateness on linguistic
processing and on linguistic encoding are investigated. A feature like
[+animate] is reflected e.g. in sortal preferences for argument roles.
An animated entity is preferably identified as the initiator of an event
and therefore assigned the agent role.

The assignment of thematic roles is part of the conceptual struc-
turing of situations, which is a complex process encompassing a va-
riety of conceptual operations. As our environment consists of a
continuous flow of activity, the perceptual and conceptual segmenta-
tion this continuation into meaningful units is a precondition to lin-
guistic encoding. This insight leads to a modification of Level's
principle of natural order, which assumes a strict correspondence of
chronological order and ordering of events.
What counts as natural ordering is different for different domains of dis-
course, and there is no general definition. Still, for certain important cases
the notion is obvious. For event structures, the natural order is the chrono-
logical order of events (Levelt 1989: 138).
That this assumption is not tenable in a strict sense has been demon-
strated in a variety of empirical investigations suggesting that, as
Zacks puts it, "events arise in the perception of observers" (Zacks
1997). Thus, for conceptualizing of event structures some additional
processes like segmentation, structuring, and selection have to be
applied prior to linearization, which transform a continuous stream
of experiences into a highly structured, often non-sequential event
structures.11
Hierarchically organized event types are sometimes held to be
stored in special sub part of the conceptual knowledge base, namely
semantic memory (cf. Kintsch 1980). Semantic memory comprises
an individual's ontological knowledge about the world at large in the
1 0
format of rather abstract types. The adjective semantic is ambigu-
ous in the given context. In psychological literature a distinction be-
tween general conceptual ontological knowledge and genuine
linguistic semantic knowledge is often either neglected, or, ignored.
In some linguistic approaches, however, semantic and conceptual
knowledge is systematically differentiated (cf. Lang, 1994 for exten-
sive arguments in favor for this distinction).

1.4. Interface in action
In the previous sections we have provided an overview of prominent
approaches to the interfacing of conceptual and linguistic representa-
tions. We have shown that from both sub-disciplines the linking be-
tween syntactic and semantic structures is either approached via
intermediate representations such as argument structure, or taken as
more or less given; e.g. in approaches that advocate quite a direct
coupling between the two as in model-theoretical theories. We have
pointed out persistent problems as how to model the different inter-
face representations or linking mechanisms and some limitations of
the respective approaches.
In the adjacent disciplines psychology and computational linguis-
tics the problem also exists but in a somewhat different fashion. In
both disciplines the processing aspect has been in greater focus as
they do not generally treat language as a formal system in its own
right. Either the overall research interest does not encompass this
aspect - as in psychology for the most part - or, is back-grounded in
the interest of building running systems.
In psychology the interfacing between different components of
the language system is for the most part regarded from the perspec-
tives of the three areas of psycholinguistic inquiry, that is acquisi-
tion, comprehension, and production. With reference to the latter
two areas the main body of research focuses on language compre-
hension, since it is of prime importance to psychological researchers
to make empirical data controllable and subject to experimental
methods. Language production research is judged less manageable in
these respects, especially concerning the production of longer strings
of language, i.e. whole utterances and texts, because it is almost im-
possible to define dependent variables in these cases. Either the ver-
balization situation has to be highly restricted,13
which then leaves
speakers no choices in how to communicate the contents in question
(and renders the whole endeavor pointless), or, the language data
become too variant to pin down the more fine grained aspects of
conceptualization and formulation.14
Thus, psycholinguistic language
production research mainly concentrates on impaired language pro-

duction (e.g. in aphasics), analyses of slips of the tongue and speech
pauses, and lexical access studies. Especially in the latter field, intri-
cate experimental paradigms have been developed to tease apart
stages during which different features of a target word become ac-
cessible: A first stage of a preverbal conceptual representation. A
second stage, during which an abstract representation of semantic
and syntactic information is retrieved (i.e. lemma selection, ibid).
And, a third stage, which eventually involves activation of the word's
phonological representation (or lexeme activation, ibid), that will
initiate articulatory encoding (cf. e.g. Jescheniak and Levelt 1994).
As becomes apparent the interface problem is thus tackled in the
transition from the conceptual component to the formulation compo-
nent, as syntactic and semantic features of the target word are acti-
vated in parallel. The utterance formulation is conceived of as being
driven via the selected lexical entries. However, the very nature of
the conceptual representation is usually not addressed as in lexical
access studies the probes for lemma and lexeme activation are either
phonetically or graphically presented word or pictures. Thus, ques-
tions of choice of open class words, collocations, connotations, and
sub-lexical relations and the like are not addressed.
This is akin to the common practice in the computer science,
where lexicalization (or lexical choice) has also become the focal
domain for a variety of sub-problems associated with the transition
from conceptual (what-to-say) to lexical representation and formula-
tion (ihow-to-say) levels (cf. Busemann 1993). Here, too, correspon-
dences between conceptual and lexical entities deviating from the
simple one-to-one pattern are not frequently encountered. In fact,
very few existing NLG systems make a distinction between concep-
tual and semantic representations in any explicit way. Typically, they
strive to reliably express their input from a well defined and limited
domain - and succeed in doing so. In parallel, the syntax-semantics
interface has been shifted into the lexicon: Most theories adhere to a
compositional semantic conception, meaning in this context the con-
struction of utterance meaning (and in the consequence utterance
structure) from the meaning of constituents and phrases. The role of
the other components has been considerable decreased in the conse-

quence and syntax is often reduced to one or two general principles.
Information concerning the categorical identity and combinatorial
constraints are projected from individual lexical entries. Lexical-
Functional Grammar (LFG, Kaplan and Bresnan 1982), Generalized
Phrase Structure Grammar (GPSG, Gazdar et al. 1985), Head
Driven Phrase Structure Grammar (HPSG, Pollard and Sag 1987)
and Unification Categorial Grammar (UCG, Zeevat et al. 1987) are
prominent examples for such monostratal and lexical theories of
grammar.15
In addition to being restricted to limited domains, existing NLG
systems encounter persistent problems in at least three fields: In the
appropriate tackling of synonyms and near-synonyms, in machine
translation and in artificial life applications. These have in common
the fact that a mere one-to-one mapping between the conceptual
level and the linguistic levels does not yield appropriate results.
The solution to these problems is for the most part sought in
modification of the system-architectures. The standard versions of
NLG systems today are modular, relying on a strictly sequential
architecture and a one-way information flow. Sequentiality and
modularity yield stability, but they also result in rigidness of the
system. The antipode to this conception is an integrated architecture,
in which knowledge at all levels acts together. Interactive
architectures are extremely flexible, albeit prone to system break-
down. Between these two extremes, we find architectures that
sequential ( 3 "C O
integrated I 1
interactive (feedback) ( ) - " ( 1
blackboard C _ J - -
I J- Η
revision-based ^ ^
Figure 1. Schemes for control of information flow (ibid)

allow for various kinds of interaction between the modules.
Interactive architectures allow for feedback processes between
modules, whereas in blackboard architectures every module has
access to common information that is shared between modules and
laid down in a mutual data structure. Revision based achitectures
allow for a limited range of feedback via monitoring components.
(For extended description of the architecture types viz. DeSmedt,
Horacek and Zock 1996).
Apart from these conceptions, there is a growing endeavor to
build hybrid models that combine advantages of different model
types. Most prominently in the revised version of Levelt's model
combines a modular architecture with interactive (connectionist)
substructures - the latter are to be found within the formulator. More
concretely the lemma-model is implemented within a spreading acti-
vation framework (WEAVER++, see Levelt, Roelofs and Meyer
1999).
In sum, the overview presented in this introduction shows that the
processing problem is tightly intertwined with, first, the kinds of
structures and processes we assume at the different stages of proc-
essing and the way we model their interaction - especially so at the
transition from conceptual/semantic to syntactic representations.
And, second, with the underlying modularity assumption, i.e. the
proposed architecture of the language faculty, which also has a
strong impact on the respective interface conceptions. The contribu-
tions in this book address these issues from various viewpoints and
theoretical backgrounds. Either they take on a model-oriented per-
spective, or, concentrate on a specific phenomenon. One phenome-
non that has currently received growing interest in the disciplines
involved is the coupling between conceptualizations of events and
their grammatical realizations. This issue is notoriously complex
(viz. paragraphs 1.1 and 1.2) as the verbalization of events varies
significantly depending on the internal features (e.g. aspectual) and
external characteristics (e.g. the chronological order) of events.
From these starting points, the current volume contributes to the
ongoing discussion about the relevance of empirical and psychologi-
cal evidence for theoretical-linguistic research and vice versa. The

book is based on the assumption that any research on human lan-
guage - even from a heuristic perspective - should include insights
into procedural aspects in the computation of a linguistic expression.
This conception has its roots in the conviction that the ways of proc-
essing data from different levels have to be reflected in the linguistic
target representation. In reverse, even though theoretical explicitness
and fine grained analyses might appear neither manageable nor de-
sirable in the implementation of NLG systems, the integration of
more findings from theoretical linguistics into computer science may
turn out to be useful in more intricate language production domains.
2. The contributions
The mediating function between concepts and grammar is ap-
proached by the contributions to this volume from three interrelated
areas of emphasis: i.) the interplay between non-linguistic and lin-
guistic information in the grammaticalization and linearization of a
preverbal message, ii.) the mapping between non-linguistic, concep-
tual event representations and the ways of verbalizing them, and iii.)
the mediating function of the lexicon in the verbalization of different
types of events. First, questions of the general architecture including
the number of levels, specific ways the information is processed on
them, and the size and the format of the interface representations is
dealt with. Here, the persistence of extra-linguistic information, its
visibility for linguistic processes, and its realization in grammar is
explored. The interplay of the several types of information involved
becomes especially apparent with the issue of event conceptualiza-
tion and verbalization, which at the same time represents a useful
basis for an application of the model assumptions developed so far.
Specifically, the question of how event concepts are stored in mem-
ory and fractionized for language processing is addressed. In this
context, a main issue to be discussed is how grammatical require-
ments determine the verbalization of event concepts and how the
interface can mediate between corresponding informational conflicts.
This thematic complex joins together the contributions of the third
section. The morpho-syntactic realization of event structural features

Mediating between concepts and language - Processing structures 2 3
and their effects on the assembly of verb complexes is projected
from principles organized in the lexicon, which are addressed in the
third group of papers. We organize the contributions according to
their main focus into the described three sections while at the same
time the interrelatedness of the issues dealt with allows for repeated
naming of one author in multiple sections. (Authors names appear in
bold letters to associate them to a respective section).
Mediating between non-linguistic and linguistic structures. The
contributions of the first section investigate the influence of different
types of extra-linguistic information on the verbalization of a linguis-
tic string. Here, affects on the linearization of a preverbal message
are of central interest. This requires a modeling of the incremental
realization of the preverbal message as well as a definition of those
meaning components which are reflected in grammar. Against this
background, FEMKE F. VAN DER MEULEN provides evidence from eye
tracking experiments that point to a close link between looking and
verbalization. Like Cummins, Gutbrod, and Weingarten, she uses
spatial configurations to elicit verbal descriptions. Her data shows
that the description of certain types of object arrays is preceded by a
preview, which interacts with the viewing times during the main pass
of the verbalization. Temporal aspect are of focal importance in the
contribution of PHILIP CUMMINS, BORIS GUTBROD, and RÜDIGER
WEINGARTEN also, where the complexity of phrasal structures is re-
lated to the time course of their production. To show also that addi-
tional conceptual information such as the size of the set of concepts
to choose from affects the verbalization of spatial configurations, the
authors provide evidence from eye-tracking and keyboard data to
underpin their hypothesis. The accessibility of conceptually differ-
ently weighted constituents is investigated by KATHY Y. VAN NICE
and RAINER DIETRICH. They disentangle extra-linguistic features
such as animacy and agentivity effects in their impact on word order
and develop a model of how this information is carried down
through the language production system. The authors thus motivate
the incremental processing models as proposed by Guhe as well as
Kempen and Harbusch by pointing to the relevance of extra-

linguistic features that become information structurally relevant dur-
ing processing. MARKUS GUHE proposes an incremental construction
of the preverbal message. He explicates how these piecemeal struc-
tures link to the underspecified semantic representations (as they are
proposed e.g. by Johannes Dölling, Veronika Ehrich, Andreas Späth,
and Ladina Tschander). Here, a critical factor is determined, namely,
the criterion that need to be fulfilled in order for a conceptual entity
to function as a legitimate increment. The incremental processing of
information on the syntactic level is central to the work of GERARD
KEMPEN and KARIN HARBUSCH, which strongly relies on experimen-
tal evidence. They apply a probabilistic method in order to model
word order phenomena in the German midfield and indicate that -
besides syntactic constraints - information structural conditions are
crucial for scrambling. Thereby they mirror the order in which the
constituents become accessible for syntactic processing during com-
putation. Considering the referential status of nominal expressions in
discourse, aspects of word order are discussed by ANDREAS SPÄTH
also. Here, the lexical principles which relate to the syntactic base
generation of lexical entries are determined. By means of these
principles - as is discussed by Veronika Ehrich and Andrea Schalley
also - the link between argument structure and word order is
accounted for where informational structural features are included
into the computational routines at work between semantics and syn-
tax. With these means presuppositions to be derived from nominal
argument phrases can be associated with a current discourse model.
From a general architectural perspective, the interaction between
grammatically visible and invisible meaning components is investi-
gated by HEIKE WIESE in her tripartite model. Drawing on empirical
evidence, she integrates insights from two-level approaches to se-
mantics with conceptual semantics. She advocates semantics as the
interface level of the conceptual and the linguistic system, where it is
a particular SEM-function that makes visible conceptual information
to the linguistic system and generates an under-specified representa-
tion.

Mediating between event conceptualization and verbalization.
Spatial and temporal configurations are to be linearized during lan-
guage production. However, while with spatial configurations the
multidimensionality of space has to be transferred onto a linear lin-
guistic sequence, with temporal relations the knowledge about the
canonical sequential ordering of events such as SOIL-WASH can be
employed for structuring the message and thus enhances processing.
This latter hypothesis is supported by the findings of ELKE VAN DER
MEER, REINHARD BEYER, HERBERT HAGENDORF, DIRK STRAUCH,
and MATTHIAS KOLBE, who show in a series of priming experiments
that the disruption of the canonical sequence of events as with WASH-
—SOIL leads to processing difficulties. The authors thus show, how
world-knowledge about events has its reflexes in linguistic event
descriptions. RALF NÜSE approaches the interrelation between event
conceptualization and event verbalization by analyzing language
specific differences between English and German speakers. By com-
paring both linguistic descriptions of visually presented events and
the corresponding eye-movements of the speakers, he comes to the
conclusion that language specific grammatical features are already at
work in the conceptualizes While she also considers the event do-
main, a modular conception is supported by MARIA MERCEDES
PINANGO. She advocates the separation between a semantic and a
syntactic module on the basis of the processing event structural
variations. She holds that utterances, in which semantic meaning is
syntactically transparent are more easily processed than those which
are compositionally enriched and thus have to be aspectually coerced
into a derived interpretation. This perspective is rejected by
JOHANNES DÖLLING. Rather than suggesting a coercion operation for
event structurally shifted expressions like John broke a cup for
weeks, he introduces a parameter which is obligatorily inserted into
the semantic representation of any verb complex. Since the parame-
ter is contextually filled, 'coercion' is reinterpreted as contextual en-
richment. The idea of enriching linguistic representations by
contextual and conceptual information is shared by MARKUS EGG
and KRISTINA STRIEGNITZ. However, in the formal realization of this
mutual understanding the two approaches differ. For one thing in the

NLG conception of Egg and Striegnitz a context-sensitive type coer-
cion operator (TC) is added to the linguistic representation only in
specific cases, namely in order to derive a well-formed syntactic
structure for expressions containing sortally coerced verb arguments,
e.g. bottle in Every bottle froze.
The mediating function of the lexicon. The lexicon is the system
where information is stored of how to relate preverbal and linguistic
structures in an economic way such that the different communicative
requirements accompanying the speech act can be met. Here, a func-
tional perspective is adopted by HEIDRUN DORGELOH and ANJA
WANNER. The authors demonstrate that the internal structure of
event concepts and their lexical argument structure, respectively, can
be made use of to meet register specific requirements. They illustrate
how the expression of certain types of events in research articles re-
late to the degrees of implicitness text producers ascribe to agentive
entities. Lexical principles controlling the derivation of nominaliza-
tions from different types of verbs are discussed by VERONIKA
EHRICH. She shows how different event structural verb types relate
to the argument structural behavior of the corresponding nominaliza-
tions. While she acknowledges that the interpretation of event nomi-
nalizations draws on conceptual knowledge, she insists that the
nominal linking rules interfacing syntax and semantics are rooted in
the grammatical system, i.e. the lexicon. Lexicon internal event en-
coding principles are treated by ANDREA C. SCHALLEY, who shares
the aspect of language comparison with Ralf Nüse. By exploring
data from Walmajarri, Kalam, and German she identifies two com-
peting lexical principles, which are derived from the language spe-
cific chunking of event concepts and determine the grammatical
alternatives of coding complex events. In the context of motion verbs
LADINA B. TSCHANDER investigates the alternation between particle
verb constructions versus prepositional phrase constructions. She
holds that conceptual conditions associated with motion and path
concepts regulate the realization of the corresponding verb com-
plexes. Thereby she accounts for the requirement that goal concepts
need to be specified in certain contexts during language production

and shows which lexical properties can adequately realize the corre-
sponding conditions.
Notes
This volume is the outcome of the workshop The Syntax-Semantics-Interface:
Linguistic Structures and Processes at the DGfS conference Language and
Cognition in March 2001. The editors' work on this volume has been com-
pleted within the projects Conceptualization processes in language production
(HA 1237-10) and Conceptual transfer of situations into verbal meaning and
the status of thematic roles (OL 101-2) of the DFG priority program Language
production and the project Semantic interfaces: copula-predicative construc-
tions at ZAS (Berlin). For constructive comments we wish thank Susan Olsen
and we are grateful for the valuable suggestions for improvement that fol-
lowed from the anonymous review process. For their competent support in the
technical realization we are emdebted to Britta Gömy, Delia Herrn, and Tho-
mas Schulz. Many thanks go to the team of Mouton De Gruyter who were ef-
ficient and helpful.
1. There exists a vast body of empirical evidence that e.g. many perceptional
processes, e.g. in visual perception, are largely autonomous of other cognitive
processes (Pylyshyn 1999).
2. The remaining three criteria relate to the biological prerequisites of modules
and Fodor holds them to be important for discerning module-generated from
learned behavior: Modules are localized, i.e. mediated by dedicated neural
structures. They obey ontogenetic and pathological universals in that they
both mature and decay in distinctive sequences.
3. Compare e.g. Frazier (1987) for a strictly modular, and e.g. Bates (1994) for a
non-modular view.
4. The syntax of a given language is semantically and pragmatically arbitrary.
For example, there are no compelling arguments outside syntax for verb-end
position in German subordinate clauses.
5. A completely different viewpoint is presented by Elman et al. (1996) and
Marslen-Wilson & Tyler (1987).
6. Here, theories of a generative character like the Government and Binding The-
ory (Chomsky 1981) focus almost exclusively on the representation of argu-
ment structure, while there is no consensus on which kind of lexical
information is to be included.
7. This conviction is shared by theoreticians outside the linking theoretical
framework (e.g. Pustejovky (1992).
8. See Härtl (2003) for discussion.

9. Friederici (1997) discusses corresponding neuro-psychological implications of
the assumption that meaning construction is achieved in two steps in language
processing.
10. This process has most prominently been accounted for in the Recognition by
Components or Geon Theory (cf. Biederman 1995). It posits that objects and
scenes are represented as an arrangement of simple, viewpoint-invariant
volumetric primitives (e.g. bricks, cylinders, wedges, and cones) termed geons
that are recognizable even if parts are occluded. Geon theory has been exten-
sively tested and can elegantly account for the fact that objects become hardly
recognizable when viewed from a highly unfamiliar perspective. A leading al-
ternative view to recognition by components is proposed by View-Based
Recognition approaches (cf. e.g. Tarr & Bülthoff 1995).
11. These processes can be characterized as follows: Segmentation of states of
affairs is the distinction of those entities that are relevant within a current con-
ceptualization, especially temporal and spatial segmentation. Structuring of
states of affairs leads to the construction of hierarchical event structures. Se-
lection singles out the subclass of available entities that are to be verbalized
(cf. Habel & Tappe 1999).
12. Following Härtl (2001: 109) we assume that during language production the
first component of the language production system, the so called conceptual-
izer, has access to the currently activated information from both the semantic
and the episodic knowledge base. Thus concrete episodic information (includ-
ing temporal and spatial specifications) can be linked to global information
about abstract event types (including abstract temporal and spatial structures)
13. Here we find a striking analogy to computational language production models:
Computer linguists have so far been forced to content themselves with very re-
stricted domains in order to build running systems, in which a coupling be-
tween the to-be-verbalized contents and language output can be guaranteed.
14. Cf. Pechmann (in print) for an overview of experimental methods in language
production research.
15. Similar trends are witnessed in linguistics, e.g. in conceptions of the genera-
tive lexicon (Pustejovsky 1995) and also in the minimalist program (Chomsky
1995,2000).
References
Aduciewicz, Kazimierz
1935 Die syntaktische Konnexität. Studio Philosophica 1, 1-27.

Anderson, John R.
1991 The adaptive nature of human categorization. Psychological Review
98: 409-429.
Baker, Mark
1997 Thematic roles and syntactic structure. In: Liliane Haegeman (ed.),
Elements of Grammar, 73-137. Dordrecht, Kluwer..
Barsalou, Lawrence W., Wenchi Yeh, Barbara J. Luka, Karen L. Olseth, Kelly S.
Mix and Ling-Ling Wu
1993 Concepts and meaning. In: Katharine Beals, Gina Cooke, David
Kathman, Karl-Erik McCullough, Sotaro Kita and David Testen (eds.),
Chicago Linguistics Society 29: Papers from the parasession on con-
ceptual representations, 23-61. University of Chicago: Chicago Lin-
guistics Society.
Barwise, Jon and John Perry
1983 Situations and Attitudes. Cambridge, MA: MIT Press.
Bates, Elizabeth
1994 Modularity, domain specificity and the development of language. In:
Discussions in Neuroscience 10: 136-149.
Bates, Elizabeth and Brian MacWhinney
1984 Cross-linguistic Studies of Sentence Processing. Cambridge:
Cambridge University Press.
Bäuerle, Rainer
1985 Das Lexikon in der kompositionalen Semantik. In: Christoph Schwar-
ze and Dieter Wunderlich (eds.), Handbuch der Lexikologie, 199-229.
Königstein, Ts.: Athenäum.
Berwick, Robert C. and Amy S. Weinberg
1984 The Grammatical Basis of Linguistic Performance. Cambridge, MA:
MIT Press.
Βickerton, Derek
1990 Language and Species. Chicago: University of Chicago Press.
Biederman, Irving
1995 Visual object recognition. In: Stephen M. Kosslyn and Daniel N.
Osherson (eds.), An Invitation to Cognitive Science, 2nd edition, Vol.
2, Visual Cognition, 121-165. Cambridge, MA: MIT Press.
Bierwisch, Manfred
1983 Semantische und konzeptionelle Repräsentationen lexikalischer Ein-
heiten. In: Rudolf Ruzicka and Wolfgang Mötsch (eds.), Untersuchun-
gen zur Semantik (Studia Grammatika 22), 61-89. Berlin: Akademie-
Verlag.
1986 On the nature of semantic form in natural language. In: Friedhart Klix
and Herbert Hagendorf (eds.), Human Memory and Cognitive Capa-
bilities - Mechanisms and Performances, 765-784. Vol. B. Amster-
dam: Elsevier.

Bierwisch, Manfred and Rob Schreuder
1992 From concepts to lexical items. Cognition 4: 23-60.
Bloom, Paul
1998 Theories of artifact categorization. Cognition 66: 87-93.
Brody, Michael and M. Rita Manzini
1988 On implicit arguments. In: Ruth Kempson (ed.), Mental
Representations: The Interfaces between Language and Reality, 105-
130. Cambridge: Cambridge University Press.
Busemann, Stephan
1993 A holistic view of lexical choice. In: Helmut Horacek (ed.), New
Concepts in Natural Language Generation: Planning, Realization, and
Systems, 302-308. London, New York: Frances Pinter.
Caramazza, Alfonso
1997 Brain and language. In: Michael S. Gazzaniga (ed.), Conversations in
the Cognitive Neurosciences, 131-151. Cambridge, MA: MIT Press.
Chomsky, Noam
1981 Lectures on Government and Binding. Dordrecht: Foris Publications.
1986 Knowledge of Language: Its Nature, Origin and Use. New York:
Praeger.
1993 A minimalist program for linguistic theory. In: Kenneth Hale and
Samuel J. Keyser (eds.), The view from Building 20, 1-52. Cambridge,
MA: MIT Press.
1995 The Minimalist Program . Cambridge, MA: MIT Press.
2000 Minimalist inquiries: The framework. In: Roger Martin, David
Michaels and Juan Uriagereka (eds.), Step by Step. Cambridge, MA:
MIT Press.
Cresswell, Maxwell J.
1978 Semantic competence. In: Franz Guenthner and M. Guenthner-Reutter
(eds.), Meaning and Translation, 9-27. London: Duckworth.
De Smedt, Koenraad, Helmut Horacek, and Michael Zock
1996 Architectures for natural language generation: Problems and perspec-
tives. In: Giovanni Adorni and Michael Zock (eds.), Trends in Natural
Language Generation: An Artificial Intelligence Perspective (Springer
lecture notes in artificial intelligence 1036), 17-46. Berlin: Springer.
Deane, Paul D.
1992 Grammar in Mind and Brain. Berlin, New York: Mouton de Gruyter.
Dölling, Johannes
1998 Semantic form and abductive fixation of parameters. Ms. Universität
Leipzig.
Dowty, David R.
1979 Word Meaning and Montague Grammar. Reidel, Dordrecht.

Ehrich, Veronika
1992 Hier undjetzt: Studien zur lokalen und temporalen Deixis im
Deutschen. Tübingen: Niemeyer
Elman, Jeffrey L., Elizabeth A. Bates, Mark H. Johnson, Annette Karmiloff-Smith,
Domenico Parisi and Kim Plunkett
1996 Rethinking Innateness. MIT Press.
Fodor, Jerry A.
1983 The Modularity of Mind. An Essay on Faculty Psychology. Cambridge,
MA: MIT Press
1998 The trouble with psychological Darwinism. London review of books.
Vol. 20, 2. [http://www.lrb.co.uk/v20n02/fodo2002.html]
Frazier, Lyn
1987a Sentence processing: A tutorial review. In: Max Coltheart (ed.),
Attention and Performance XII: The Psychology of Reading. Hillsdale:
Lawrence Erlbaum.
1987b Theorie of sentence processing. In: Jay L. Garfield (ed.), Modularity in
Knowledge Representation and Natural Language Understanding.
Cambridge, MA: MIT Press.
Frazier, Lyn, Charles Clifton and Janet Randall
1983 Filling gaps: Decision principles and structure in sentence comprehen-
sion. Cognition 13: 187-222.
Friederici, Angela D.
1997 Neurophysiological aspects of language processing. Clinical
Neuroscience 4: 64-72.
Gärdenfors, Peter
2000 Conceptual Spaces. Bradford Books. Cambridge, MA: MIT Press
Gazdar, Gerald, Ewan Klein, Geoffrey Pullum and Ivan Sag
1985 Generalized Phrase Structure Grammar. Oxford: Basil Blackwell.
Grimshaw, Jane
1979 Complement selection and the lexicon. Linguistic Inquiry 10: 279-
326.
1990 Argument Structure. Cambridge, MA: MIT Press.
1993 Semantic structure and semantic content in lexical representation. Un-
published ms., Rutgers University, New Brunswick, NJ.
Grimshaw, Jane and Sten Vikner
1993 Obligatory adjuncts and the structure of events. In: Eric Reuland and
Werner Abraham (eds.), Knowledge and Language: Vol. II: Lexical
and Conceptual Structure, 145-159. Dordrecht: Kluwer.
Habel, Christopher and Heike Tappe
1999 Processes of segmentation and linearization in describing events. In:
Ralf Klabunde and Christiane v. Stutterheim (eds.), Representations
and processes in language production, 117-153. Wiesbaden: Deut-
scher Universitätsverlag.

Härtl, Holden
2001 CA USE und CHANGE: Thematische Relationen und Ereignisstruktu-
ren in Konzeptualisierung und Grammatikalisierung. (Studia Gram-
matica 50). Berlin: Akademie-Verlag.
2003 Conceptual and grammatical characteristics of argument alternations:
The case of decausative verbs. Appears in: Linguistics 41(5).
Jackendoff, Ray
1972 Semantic Interpretation in Generative Grammar. Cambridge, MA:
MIT Press.
1990 Semantic Structures. Cambridge, MA: MIT Press
1992 Languages of the Mind. Cambridge, MA: MIT Press.
1997 The Architecture of the Language Faculty. Cambridge, MA: MIT
Press.
Jescheniak, Jörg D. and Willem J. M. Levelt
1994 Word frequency in speech production: Retrieval of syntactic informa-
tion and phonological form. Journal of Experimental Psychology:
Learning, Memory and Cognition 20: 824—843.
Kaplan, David
1977 The logic of demonstratives. Journal of Philosophical Logic 8: 81-98.
Kaplan, Ronald M. and Joan Bresnan
1982 Lexical-functional grammar: A formal system for grammatical repre-
sentation. In: Joan Bresnan (ed.), The Mental Representation of
Grammatical Relations. Cambridge, MA: MIT Press.
Karmiloff-Smith, Annette
1992 Beyond Modularity: A Developmental Perspective on Cognitive Sci-
ence. Cambridge, MA: MIT Press.
Katz, Jerrold J. and Jerry A. Fodor
1963 The structure of a semantic theory. Language 39: 170-210.
Koenig, Jean-Pierre and Gail Mauner
1999 A-definites and the semantics of implicit arguments. Journal of
Semantics 16: 207-236.
Lakoff, George
1991 Metaphor and war: The metaphor system used to justify war in the
Gulf. Distributed via electronic bulletin boards. Journal of Urban and
Cultural Studies 2.
Lang, Ewald
1994 Semantische und konzeptuelle Struktur: Unterschneidung und Über-
schneidung. In: Monika Schwarz (ed.), Kognitive Semantik: Ergebnis-
se, Probleme, Perspektiven, 25-41. Tübingen: Narr.
Langacker, Ronald W.
1987 Foundations of Cognitive Grammar I: Theoretical Prerequisites. Stan-
ford: Stanford University Press.

1990 Settings, participants, and grammatical relations. In: S. Tsohatzidis
(ed.), Meanings and Prototypes: Studies in Linguistic Categorization,
213-238. New York: Routledge, Chapman and Hall.
1991 Foundations of Cognitive Grammar II: Descriptive Applications.
Stanford: Stanford University Press.
Levelt, Willem J. M.
1989 Speaking: From Intention to Articulation. Cambridge, MA, London,
UK: MIT Press.
Levelt, Willem J. M., Ardi Roelofs and Antje S. Meyer
1999 A theory of lexical access in speech production. Behavioral and Brain
Sciences 22: 1-75.
Levinson, Stephen C.
1996 Frames of reference and Molyneaux's question: Cross-linguistic evi-
dence. In: Paul Bloom, Mary A. Peterson, Lynn Nadel and Merrill F.
Garrett (eds.), Language and Space, 109-169. MIT Press.
Levin, Beth and Malka Rappaport Hovav
1995 Unaccusativity: At the Syntax-Lexical Semantics Interface. (Linguistic
Inquiry Monograph 26). Cambridge, MA: MIT Press.
Lewis, David
1972 Psychophysical and theoretical identifications. Australasian Journal of
Philosphy 50: 249-258.
Marcus, Mitchell P.
1982 Consequences of functional deficits in a parsing model: Implications
for Broca's aphasia. In: Michael A. Arbib, David Caplan, John C. Mar-
shall (eds.), Neural Models of Language Processes (Perspectives in
Neurolinguistics, Neuropsychology, and Psycholinguistics), 115-134.
New York: Academic Press.
Marshall, John C.
1984 Multiple perspectives on modularity. Cognition 17: 209-242.
Marslen-Wilson, William D. and Lorraine K. Tyler
1987 Against modularity. In: Jay L. Garfield (ed.), Modularity in
Knowledge Representation and Natural-Language Understanding.
Cambridge, MA: MIT Press.
McCawley, James D.
1971 Tense and time reference in English. In: Charles Fillmore and D.
Terence Langendoen (eds.), Studies in Linguistic Semantics, 97-113.
New York: Holt Rinehart.
McShane, Marjorie and Ron Zacharski
2001 Modularity in knowledge elicitation and language processing. Pro-
ceedings of the Third Annual High Desert Linguistics Conference.
University of New Mexico, Albuquerque NM, April 7-9, 2000. 93-
104.

Medin, Douglas L.
1989 Concepts and conceptual structure. American Psychologist 44: 1469-
1481.
Meisel, Jürgen M. (ed.)
1990 Two First Languages: Early Grammatical Development in Bilingual
Children. Dordrecht: Foris.
Mohanan, Tara and Lionel Wee (eds.)
1999 Grammatical Semantics: Evidence for Structure in Meaning. Stanford,
CA: CSLI Lecture Notes.
Montague, Richard
1970 Universal grammar. Theoria 36, 373-398.
Montague, Richard
1973 The proper treatment of quantification in ordinary English. In: Jaakko
Hintikka, Julius M. Moravcsik and Patrick Suppes (eds.), Approaches
to Natural-Language: Proceedings of the 1970 Stanford Workshop on
Grammar and Semantics, 221-242. Dordrecht: Reidel. [Reprinted in
Montague (1974). 247-270.].
Müller, Ralph-Axel
1996 Innateness, autonomy, universality? Neurobiological approaches to
language. Behavioral and Brian Sciences 19: 611-675.
Newmeyer, Frederick J.
1998 Language Form and Language Function. Cambridge, MA: MIT Press.
O'Bryan, Erin, Raffaella Folli, Heidi Harley and Thomas Bever
in prep. The Role of Event Structure in Language Comprehension,
http://www.ling-phil.ox.ac.uk/research/folli/research_files/amlap.doc
Olsen, Susan
1998 Prädikative Argumente syntaktischer und lexikalischer Köpfe: Parti-
kelverben im Deutschen und Englischen. Folia Linguistica 31: 301—
329.
Partee, Barbara
1975 Montague grammar and transformational grammar. Linguistic Inquiry
6(2): 203-300.
Pechmann, Thomas
in print Experimentelle Methoden in der Sprachproduktionsforschung. In:
Theo Herrmann and Joachim Grabowski (eds.), Enzyklopädie der Psy-
chologie, Band C III/1: Sprachproduktion. Göttingen: Verlag für Psy-
chologie.
Pollard, Carl and Ivan Sag
1987 An Information-based Approach to Syntax and Semantics: Vol. 1 Fun-
damentals. Stanford, CA: CSLI Lecture Notes.
Pustejovsky, James
1995 The Generative Lexicon. Cambridge, MA: MIT Press.

Pylyshyn, Zenon W.
1999 Is vision continuous with cognition? The case for cognitive
impenetrability of visual perception. Behavioral and Brain Sciences
22(3): 341—423.
Rappaport, Malka and Beth Levin
1988 What to do with theta-roles. In: Wendy Wilkins (ed.), Syntax and
Semantics 21: Thematic Relations, 7-36. New York, NY: Academic
Press.
Roeper, Thomas
1987 Implicit arguments and the head-complement relation. Linguistic
Inquiry 18: 267-310.
Saussure, Ferdinand de
1916 Cours de linguistique generale. In: Charles Bally and Albert
Sechehaye (eds.), Paris: Payot.
Smolensky, Paul
1988 On the proper treatment of connectionism. Behavioral and Brain Sci-
ences 11: 1-74.
Stillings, Neil Α.; Steven E. Weisler, Christopher H. Chase, Mark H. Feinstein, Jay
L. Garfield and Edwina L. Rissland
1998 Cognitive Science: An Introduction (2nd ed.). Cambridge, MA: MIT
Press.
Stromswold, Karin
1992 Leamability and the acquisition of auxiliary and copula be. In:
ESCOL '91. Columbus State University.
Tappe, Heike
1999 Der Spracherwerb bei Corpus-Callosum-Agenesie: Eine explorative
Studie. Tübingen: Narr.
2000 Perspektivenwahl in Beschreibungen dynamischer und statischer We-
geskizzen. In: Christopher Habel and Christiane v. Stutterheim (eds.),
Räumliche Konzepte und sprachliche Strukturen, 69-97. Tübingen:
Niemeyer.
Tarsky, Alfred
1977 Einföhrung in die Mathematische Logik. Göttingen: Vandenhoeck and
Ruprecht.
Tarr, Michael J. and Heinrich H. Bülthoff
1995 Is human object recognition better described by geon structural de-
scriptions or by multiple views? Comment on Biederman and
Gerhardstein (1993). Journal of Experimental Psychology: Human
Perception and Performance. 21(6): 1494-1505.
Taylor, Holly A. and Barbara Tversky
1996 Perspective in spatial descriptions. Journal of Memory and Language
35,371-391.

Tenny, Carol
1992 The Aspectual Interface Hypothesis. In: Ivan A. Sag and Anna
Szabolcsi (eds.), Lexical Matters (CSLI lecture notes 24), 1-27. Palo
Alto: Stanford University Press.
1994 Aspectual Roles and the Syntax-Semantics Interface. (Studies in Lin-
guistics and Philosophy 52). Dordrecht: Kluwer.
Tomasello, Michael
2000a Do young children have adult syntactic competence? Cognition 74:
209-253.
2000b The item based nature of children's early syntactic development.
Trends in Cognitive Sciences 4: 156-163.
van Hout, Angeliek
1996 Event semantics of verb frame alternations: A case study of Dutch and
its acquisition. Ph.D. dissertation, Tilburg University.
Wilder, Chris and Damir Cavar
1994 Long head movement? Verb movement and cliticzation in Croatian.
Lingua 93: 1-58
Wunderlich, Dieter
1997 CAUSE and the structure of verbs. Linguistic Inquiry 28(1): 27-68.
Zacks, Jeffrey M.
1997 Seeing the structure in events. Ms., Stanford University.
Zeevat, Henk, Ewan Klein and Jo Calder
1987 Unification categorial grammar. In: Nicholas Haddock, Ewan Klein
and Glyn Morrill (eds.), Categorial Grammar, Unification Grammar,
and Parsing. University of Edinburgh, Scotland.

Mediating between
non-linguistic and linguistic structures

Another Random Scribd Document
with Unrelated Content

FORTUNES WON AND
LOST.
The first movement for utilizing this vast oil product was made, in
1854, by two New Yorkers, who organized a company, and secured
the right to a certain spring on Oil Creek; but they made no progress
until three years later, when Messrs. Bowditch and Drake, of New
Haven, undertook to search for oil. In the winter of 1858 and ‘9,
Colonel E. S. Drake completed arrangements for boring into the rock
below the bed of the creek. On the 26th of August, 1859, oil was
found at a depth of seventy-one feet. The drills sank into a cavity in
the rock, and the oil rose to the surface. By means of a pump, four
hundred gallons were obtained per day, and a larger pump being
introduced, the supply reached one thousand gallons daily. This was
the beginning of the borings for oil in that region. Every spot where
oil was found, or was likely to be found, was carefully examined, and
a great many wells were put down. Up and down the banks of Oil
Creek derricks were erected and wells were sunk, and in a year or
two the banks of the stream looked as if their natural product had
been derricks rather than trees. The ground was perforated like a
sieve, and if the holes had been a few feet, instead of a few inches,
in diameter, it would have been dangerous walking round there for
fear of tumbling through. The original depth of seventy-one feet was
found insufficient, and the borings were frequently conducted to a
depth of several hundred feet. I believe one well was sunk over two
thousand feet, and a great many wells exceeded a thousand. Many
of them never produced oil, and the man who had risked his money
to bore these wells saw it vanish without affording anything in
return.
A great many stories are told of fortunes
made and lost in boring for oil. In some
cases men just narrowly missed success,
and in others they obtained their success by accident. A story is told
of some men who had secured a locality and sunk their drills to a
depth of nearly a thousand feet. All their money was gone, and they
knew not where to obtain more. There were no indications of oil,
their machinery was mortgaged, and the sheriff stood by to secure
it. They were about to abandon work; it was near the close of the

day, and they had no credit and no means to continue work on the
following morning. One of the men proposed to quit about four
o’clock in the afternoon. “No,” said the other; “let us die game, and
put the machine through till sunset.”
He tore away a piece of the timber supporting the derrick, and
threw it into the furnace to give additional speed to the engine. Just
as the sun was beginning to dip behind the western hills, the drill
suddenly sunk several feet. It was withdrawn from the rock, and a
column of oil mixed with salt water followed it. They had “struck oil,”
and were saved.
In another instance a company was formed, and had drilled a
dozen wells, but without success. Their capital was nearly gone, and
they were working on a well which, if unsuccessful, would prove
their ruin. Just as they had expended almost their last dollar, and
were within twenty-four hours of suspending, they found oil in
abundant quantities, and were saved from ruin.
There are many instances of men searching for oil, boring their
wells to a considerable depth, and abandoning them in consequence
of the exhaustion of their money, and the discouraging prospects.
After abandoning their work, others took possession of the places,
and in a few days, sometimes in a few hours, opened wells of great
value.
In the oil regions I was once told a story of two men who had
been at work a long time, but could get no oil. Their money was
exhausted, and they became discouraged. When they had expended
their last dollar, and mortgaged everything, they stepped aside and
made way for their creditors. As they surrendered their machinery,
one of them said,—
“Let us clear out of this place, and go to work by the day until we
can get enough to try it again.”
“Hold on,” said the other; “let us sit down and see these fellows
work. We will stay a little while, and see if they get along as fast as

A LUCKY STROKE.
SALTING A WELL.
we did.”
So the two men remained, mainly for the reason they did not
know what else to do.
The new comers drilled away at the
well, which was already several hundred
feet deep, and in half an hour after they began working they found
oil. When the tools were withdrawn, the well began flowing a
hundred or more barrels per day. Imagine the disgust of the former
owners!
It was the same in the oil regions with regard to disappointments
that it has been in California and other countries containing mineral
treasures. A case like the one just described is almost an exact
parallel of a case in California, where two men, working a week or
more on a claim where they hardly made money enough to pay their
expenses, abandoned it in disgust. Two others stepped in, and on
the very day they took possession, found a lump of gold worth
several thousands of dollars. In another instance some Americans
abandoned a claim, which was immediately occupied by half a dozen
Chinese. The Chinese found a rich deposit of gold within six inches
of where one of the Americans had abandoned the use of his pick
and shovel.
Petroleum wells can be “salted” or
“baited,” just as gold or other mines can
be salted, and in the early days of the oil fever, the baiting of
petroleum wells was by no means an uncommon thing. Sometimes it
would be done by one of the owners of a well in order to defraud
other owners. For instance, Smith and Brown have entered into
partnership to put down a well. They join their money together, buy
the necessary drills and machinery, and go to work. The well is down
one or two hundred feet. Smith gets tired of it. He knows that Brown
has more money, and so thinks that he will sell out. While Brown is
asleep, Smith gets a barrel or so of petroleum, and pours it into the
well. Next morning, when they go to work, the condition of the hole

A NEAT SWINDLE.
is tested as usual, and of course there are indications of petroleum.
If a barrel has been poured into the hole it is filled for quite a long
distance. Smith has taken care to be away at the time, and appears
in perfect ignorance. If Brown is honest he will tell Smith, on his
reappearance, of the rich supply they have found; but the chances
are two to one that Brown will say nothing, except to suggest
carelessly that the well is not very promising, and ask Smith what he
will give for his share. Smith says, with equal carelessness, “I don’t
want to buy, but I will sell my interest for three thousand dollars.”
Perhaps he puts it at a higher figure. He knows the length of
Brown’s purse, and goes for its contents. The result is, that Brown
secretly chuckles over his speculation, and buys the well.
Smith goes on his way rejoicing, and Brown, still more rejoicing,
stays where he is. He knows that a few inches more of depth to the
well will yield abundant oil, and he works away very earnestly; but
somehow he keeps on drilling for a long time, and at last awakens to
the consciousness that he has been sold.
A great many petroleum wells have been salted and sold in this
way, but it sometimes happens that the would-be swindler gets the
worst of his bargain. I knew one case, in 1863, where a man baited
a well in the above way, and sold it. He laughed that evening over
his sharp trick; but he laughed less the next morning, when he
passed the well and saw that the tools had been withdrawn, and the
well was flowing at the rate of three hundred barrels a day. A few
hours after the purchasers entered upon their work, they struck oil
and were happy.
A trick that has been practised in the oil
regions to some extent is to convert a well
which has no oil in it into a genuine flowing well. I have known this
to be done by conducting a pipe underground from a tank at a
genuine well a few hundred yards away. The pipe opens into the
baited well, and it can readily be seen that with a good “head” on
the pipe the well will be a perfect flowing well, to all intents and

FRAUDULENT OIL
COMPANIES.
purposes. Men are engaged in barrelling the substance, and a visitor
can see with his own eyes the amount of the yield. If he wants to
buy a well, nobody has any great desire to sell, and he may have
difficulty in buying the whole thing outright; but he can get an
interest in it for a comparatively low figure. Sometimes he may buy
one man’s interest, and then another man’s, and he thinks he has
struck a very fine bargain. But during the night, after his purchase,
the oil ceases flowing, and he finds that his property is worthless.
Another swindle of the same sort is to have a tank filled with oil,
and a pipe run through one of its supporting posts, and under
ground into the well. The pumping machinery is kept at work, and it
may be pumping, say, at the rate of one hundred barrels a day. But
all the time that the pump is working, the oil is running into the well,
and it may run in and be pumped out again and again. The
operation is a simple one, and well calculated to deceive.
A great many petroleum companies
were organized at one time, which had no
existence beyond the paper one that they
had in New York and other cities. Some of these companies gave
most brilliant promises. I remember one which printed a flaming
prospectus, and announced that there was room on its territory for
three thousand first-class wells. No one could doubt the truth of this
assertion, but its territory happened to be on the top of a mountain,
where three hundred thousand wells might have been sunk without
finding a drop of oil. The projectors of this concern sold a great deal
of stock, but I believe they never declared a dividend of a single
dollar, or even took the trouble to sink a well. Their money was
made by defrauding their patrons rather than by doing any work in
an honest way. Millions of dollars were sunk in oil speculations
whose investors never obtained any return whatever. The public
heard of the wells that yielded enormously, but they never heard of
the thousands of wells that never amounted to anything.
So great was the rage for oil speculation during the height of the
fever, that a well would be sunk where there was the least chance or

THE SPRING THAT
FLOWED WHISKEY.
prospect of obtaining oil. Suppose a man found a spring of pure
water; he might pour a gallon or so of oil on the surface, and then
carelessly, and with apparent innocence, lead a stranger to the
vicinity. The stranger soon smells the oil, examines the water, and
buys the spring at a high price.
One day a farmer broke a kerosene lamp in his cellar. A few hours
later he admitted a stranger who wanted to buy some potatoes. The
stranger discovered the oil, forgot about the potatoes, and
immediately opened negotiations for buying the house and the land
on which it stood. He paid about three times as much as they were
worth, and the farmer went away happy.
A man, who thought crude petroleum a good remedy for freckles,
one day bathed his face in that article, and lay down to sleep. As he
tells the story, he was waked in half an hour by a New York
speculator who was trying to sink a shaft into his ear.
A story is told in California of a man
owning a farm which he wanted to sell. He
had heard of the petroleum dodge, and
thought he would try the same plan in another way. So one day,
when a lot of speculators from San Francisco were at his house, he
poured a gallon of whiskey into a small spring, and then led the
speculators in that direction. The farmer spoke of the spring, said
that he made no use of it, as he had an abundance of water near his
house. He had never observed the spring except to remark its
peculiar color. He roused the curiosity of the strangers so that one of
them tasted the water, winked at his neighbor, and stepped aside.
Before night the farmer had sold his place at a high price, and the
speculators had organized a company for supplying the California
market with an excellent article of whiskey cocktail. But somehow
their enterprise never succeeded.
The immense fortunes made from petroleum speculations were
almost marvellous; a man might be poor to-day and worth a million
dollars to-morrow. In the morning he could not raise enough money

THE MILLIONNAIRE
YOUTH.
to buy a breakfast, and at noon his credit would be good for the
purchase of a first-class steamship. A man might be working as a
day laborer this week, and his wife would be taking in washing at a
dollar a dozen. Six days later he would be a millionnaire clad in
broadcloth and fine linen, and wearing a diamond like a calcium
light, while his wife would be arrayed in silks of the most costly
character, and wearing them as uneasily as a bull-dog wears a pair
of trousers tied around his neck.
A good story is told of a woman one day selecting some diamonds
in a jewelry store on Broadway. Two other women were standing
near and observing her motions. One of them suggested to her
friend, “Evidently shoddy.”
The diamond purchaser raised her eyes for a minute, and said,
“No, madam; petroleum.”
A great many stories are told of a youth in the oil regions who was
brought up on a farm, and who, for a year or more, after the
outbreak of the oil fever, was driving a team at fifteen dollars a
month. He had a grandmother, as most young men have, but she
was unlike a great many grandmothers, as she was enormously rich.
She owned a large farm, and leased it to speculators who wished to
search for oil. She always stipulated for half the oil, and her farm
was so productive that she had a magnificent income, and
accumulated money at a very rapid rate. A common report was, that
she had eleven barrels and four trunks full of greenbacks.
One day she did as all good
grandmothers do,—she died. The youth,
whom I will call John, as that was half his
name, became heir to her vast estate. He dropped into two millions
of cash, and into the farm, which yielded about two thousand dollars
a day. He had never had so much money before in all his life. Ox-
driving at the compensation he received would require a long time
for the accumulation of such a fortune.

WASTING HIS
SUBSTANCE.
He thought the matter over, and determined to have a good time.
He engaged several youths of his acquaintance to assist him in
wasting his substance in riotous living. The party went first to
Cleveland. At the railway station they had some dispute about a
carriage, and so John bought a carriage to take them to their hotel.
When he reached the hotel he concluded that that was not the kind
of carriage he wanted, and so gave it away. He secured all the best
rooms in the house, ordered the best supper the proprietor could
furnish, and the party went to bed on the floor as drunk as a
quartette of badgers. They rose the next morning with very large
heads on their shoulders, and were occupied during the forenoon in
removing their Mansard roofs by means of soda water and cocktails.
John sent for the best team in Cleveland, and obtained a four-
horse one, with a carriage gorgeous enough for a third-rate emperor.
He picked out one of the drivers round the front of the hotel, told
him they were going to stay in Cleveland a few days, and if this
driver would take the team and drive them round during their stay,
he should have the whole concern, at their departure, for his
trouble.
John next proposed to charter a grog-shop, and another
institution which shall be nameless, for the exclusive use of himself
and friends during their stay. They made things lively for a few days,
and then left for Philadelphia by way of Buffalo.
They stopped at Niagara Falls, and proposed hiring a boat-load of
people to be sent over the falls for their amusement; but, somehow,
they could not find anybody willing to make the jump. John wanted
to buy the Falls and run them as a private show, but he changed his
mind and continued his journey.
In Philadelphia, and subsequently in
New York, the party was guilty of various
extravagances, and sometimes displayed
absolute ingenuity in getting rid of their money. On one occasion
they treated a party of fifty or more street laborers to champagne,

GETTING UP A PARTY.
filling each of them up to his chin, and sending them home blind
drunk. They bought horses and carriages to give away next day.
They chartered hotels and other public resorts for their exclusive
occupation. They used to give away ten-dollar bills, and sometimes
hundred-dollar bills, as gratuities to servants.
John seemed to be troubled to know what to do with his money,
and it gave him more anxiety than he was ever blessed with during
the days of his ox-driving experience. I believe he died after a year
or so of this new life. It was too much for him; he could endure
poverty, but he could not enjoy or endure such an accumulation of
wealth.
There was a case similar to his of a young man growing suddenly
rich through petroleum, who started on a riotous career, and
managed to get heavily in debt. The wells gave out, and left him
without money, and no prospect of obtaining any. In a year from the
time of his becoming so suddenly wealthy, he was at work again as
a day laborer, and meditating upon the uncertainties of life in the oil
regions.
On one occasion an oil speculator came to New York with fifty
thousand dollars or more in cash, and claiming that he had a flowing
well yielding two hundred barrels a day. In less than a fortnight he
had gambled away his money, sold his wells, and the last I saw of
him he was on his way to the station-house for default of paying the
amount of his hotel bill. He was kept there a short time, and then
released. I believe the hotel never received anything from him.
A great many extravagances have been committed by the
petroleum aristocracy. Persons suddenly raised from poverty to
affluence are nearly always anxious to effect an entrance into
society. They take fine houses, and sometimes they manage to get
people of repute to visit them, though not often.
Three or four years ago a family that
had suddenly grown rich determined to

give a party that should introduce them to society. They made
preparations, and sent out a great many cards of invitation. They
ignored their former acquaintances altogether. They selected the
names of their guests from the City Directory, taking those that were
prominent in the social world. They even pretended to an aristocratic
descent, and I believe their card of invitation bore a crest of some
sort or other.
The evening of the entertainment came. Madame, almost
smothered in silks, with a large amount of store hair, and decked
with diamonds enough to set up a jewelry store, was all ready to
receive her guests. The daughters were in their best, and expected
to make a dozen conquests apiece in the course of the evening. A
magnificent supper had been prepared, and a troupe of servants
were awaiting the commencement of their duties. Eight o’clock was
the hour fixed for the party.
At eight o’clock there was not a guest in the house. “Surely,” said
Madame, “they will be here very soon.” Half past eight o’clock came.
Nobody. Nine o’clock. Nobody. Half past nine. Nobody; and then ten
o’clock, and still Nobody. It was then the great truth stood revealed
that the party was a failure.
The servants, who had been standing about with their tongues in
their cheeks, were commissioned to eat what they could of the
gorgeous banquet, and the aspirants to social honors smothered
their sorrow, and made no more attempts, for that season at least,
to get into society.

PLACES FOR WINE UNDER
GROUND.
XXIII.
WINE AND BEER CELLARS.
WINE CELLARS.—HOW THEY ARE MADE.—PLACES FOR STORING BEER.—THEIR
EXTENT.—THE GREATEST WINE CASK IN THE WORLD.—ITS CAPACITY.—
PECULIARITIES OF WINE AND BEER VAULTS.—VISITING A CELLAR IN
POLAND.—CURIOUS SIGHTS.—THE ANTIQUITY OF THE BOTTLES.—WHAT A
VISITOR DID.—THE RESULT OF TOO MUCH WINE.—A DANGEROUS BRIDGE.
A German resident of New York, engaged in the manufacture of
beer, visited the excavations at Hallett’s Point, near the upper end of
Manhattan Island, and, on viewing the large space which had been
dug out of the solid rock, exclaimed, “What a capital place for
storing lager beer.” Many a wine and beer manufacturer has made
the same remark on visiting the Mammoth Cave, or other huge
caverns. The best places for storing malt or vinous liquors are under
ground, for the reason that an equal temperature can be maintained
at all times; summer’s heat and winter’s cold make but very little
change of the thermometer in the depths of the earth.
In various parts of the world, particularly
in Europe, there are vast underground
spaces specially designed for the storage
of wine, beer, and similar beverages. Nearly all these articles require
to be kept some time before they are fit for use; especially is this the
case with wines, some of which improve steadily during a year, or for
ten, or twenty, fifty, or it may be for a hundred, or five hundred
years. Some of the wine cellars of Europe have been hewn out of
the solid rock, or dug out of the solid earth, at vast expense, for the
simple purpose of storage. Other wine cellars were, originally,
quarries, or mines; and after they had been abandoned by the
miners, they were taken up by the wine and beer manufacturers,
and adapted to their present uses. The same is the case in America.

Reference is made elsewhere to the cellars of Dubuque, Iowa, which
are nothing more nor less than exhausted lead mines. At several
places on the Mississippi and Missouri Rivers there are cellars which
originally were quarries or mines. Their natural treasures were taken
from them, and they are now filled with artificial ones.
In California, particularly in the Sonoma Valley, are some wine
cellars which have been dug out of the rock for no other purpose
than for that of storage. Some years ago I visited one of these
establishments with a small party, and the proprietor, in order to give
us an idea of the temperature, shut us up a little while, and left us to
ourselves. The place was not cold, but it was cool compared with the
outer atmosphere, and we very soon began to sneeze. Had we been
kept there for any length of time, I suspect that we would have had
sore throats and all that sort of thing; but they were prevented by
the select assortment of liquids which the wine manufacturer
supplied to us with such liberality that some of his visitors’ legs
became very much entangled, and refused to perform the duty
usually required of them.
All through Europe, and particularly in France and Germany, there
are cellars of great extent. The wine makers of France and Germany
are able to store away thousands of casks, and other thousands of
bottles, every year without any difficulty. The same is the case with
the beer makers of North and South Germany, particularly in the
vicinity of Munich and Vienna. There is one wine cellar on the
Moselle, which is said to be capable of containing a million bottles
and twenty or thirty thousand casks of wine at one time, and I have
heard of one wine cellar even larger than this. The capacity of the
beer vaults of Munich is, I think, greater than that of the German
and French wine vaults. It is certain that a storage capacity sufficient
to supply the annual consumption of beer in Munich, Vienna, or
Berlin, must approach the dimensions of a small city. It is well
known that the average German can get outside of a great quantity
of beer in the course of twelve months. As an illustration, I may
mention that the day before writing this paragraph I was told of a

FAMOUS BEER DRINKERS.
THE GREATEST WINE
CASK.
strike among some German laborers in an establishment near New
York. Their strike was not for wages, but for beer. They were
satisfied with the pay they received, but not with the quantity of
beer furnished to them. Their employer allowed them two five-gallon
kegs daily for every three men, and in their strike they demanded a
daily keg of beer per man. They said that two thirds of a five-gallon
keg were not sufficient, but they would manage to get along with
five gallons each per day. The employer agreed with them, and they
resumed work as soon as he consented to their demand.
It is on record that one individual
German drank one hundred and fifty
glasses of beer per day, and I believe there was an instance in
Cincinnati, a few years ago, where a German consumed, on a wager,
one hundred and eighty-eight glasses between sunrise and sunset of
a summer’s day. It is not fair to take these ambulatory beer casks as
an indication of the drinking abilities of the Teutons, but it may
safely be assumed that an ordinary community of Germans can get
outside of an average of twenty glasses a day per man without
feeling it.
It is not my province to describe the
process of making beer or wine, as the
work is mainly performed above ground,
but simply to allude to the space where these beverages are stored.
I have visited a fair proportion of them in various parts of the world,
and they are all pretty much alike. They are simply large vaults or
caves, sometimes arched over to prevent the falling in of the earth,
while in other cases they are cut out of the solid rock, and require
no arching. Sometimes a wine cellar will consist of a single vault,
with regular pillars or arches sustaining its roof, while in other cases
there will be a great many galleries, or tunnels, running off in
different directions. Sometimes the casks containing the wine or
beer will be of a size that will permit of their being rolled about,
while in other cases the casks or tuns will be so large that they
always remain stationary, and are filled and emptied without being

A WINE CELLAR IN
WARSAW.
moved from their places. An example of this is the celebrated tun of
Heidelberg, constructed in 1751, and capable of containing forty-
nine thousand gallons. It has been filled but two or three times since
its construction, and the process of filling occupied on each occasion
two or three weeks. It is sufficiently large to allow the erection of a
ball-room upon it, and several festivals and dances have been held
there. It is the largest cask which has ever been made, or probably
ever will be made.
The preparation and preservation of wine require great care, and,
above all things, an even temperature. Many a cask of wine has
been spoiled by being kept too hot or too cold; and this is one
reason why the preference is shown by wine makers for
underground places of storage. Apart from this fact is the saving
that can be made by utilizing the space under the earth where the
surface is of great value.
As before stated, a visit to one wine cellar is very much like a visit
to another. The stranger is led or guided among rows of casks and
bottles, and sometimes his underground journey will amount to a
mile, or two or three miles, of linear distance. He wonders how the
demand can be so great for this material, just as a countryman
wonders, as he walks through the market of a large city, how all the
beef, pork, and mutton can find purchasers. He may go through a
market and think the supply exceeds the demand, just as when he
walks the streets for an hour or two, and sees the crowds of people,
he will wonder where all this mass of humanity can find sufficient
food. In the same way a person unfamiliar with the business may
have alternate surprises about the supply and consumption of wine.
One of the first wine cellars which I
visited in Europe was in the famous city of
Warsaw, Poland. I had entered Europe by
the back door, as it were, coming from Asia over the Ural Mountains;
and consequently the first ancient city I found where there was any
wine trade of significance was Warsaw. A travelling friend and myself
were under the guidance of an officer serving on the staff of the

governor of Poland, and while pointing out the curiosities of the city,
he suggested taking us to one of the oldest wine cellars in Europe. I
think he said there were a few, but only a few, which had greater
antiquity.
Our party was small,—only three of us altogether,—and we drove
in a single carriage to a very unattractive place in the Jews’ quarter
of Warsaw. We entered a narrow and rickety-looking building, which
gave no promise of the wealth stored away beneath it. The officer
was acquainted with the proprietor of the place, so that we easily
obtained permission and escort for our underground journey. The
proprietor himself took charge of us, and was accompanied by a
servant to assist in showing us round, and possibly to see that we
did not stow away in our pockets any of the valuable bottles in the
cellar.
We descended a narrow stairway, so narrow, in fact, that we went
singly, and so low that we were obliged to stoop to avoid hitting our
heads. The place was hewn out of the rock on which Warsaw is
built, and it was arched over to sustain the weight resting upon it.
Reaching the floor of the cellar, we were first led between rows of
casks, and the ages of the casks were stated as we walked among
them. One was pointed out that had been in the cellar thirty years,
and another that had been there two or three times as long. They
were covered with dust and cobwebs, and looked as if good for a
much longer stay. Over our heads we could hear the rumbling of
carriages in the streets, just as one can hear the carriages in
exploring the ruins of Herculaneum.
Cask after cask was pointed out, until our eyes were wearied, and
we were then taken to the old cellar where the bottles were stored.
Our guide explained that the cellar we had just visited was a
modern one, only two hundred and sixty years old. The old cellar, he
said, was made in the days when Poland was a kingdom, and more
powerful by far than the now great Muscovite empire. I do not
remember positively the age he gave it, but I think it was some nine

OLD BOTTLES OF WINE.
hundred or a thousand years old. I was too busy looking among the
bottles to take particular notice of what he said, and am not willing
to trust too much to my memory, especially on the occasion of
visiting a cellar like this. The real interest of the place began when
we entered the locality where the bottles were stored. Here were
little shelves—I say little, though many of them were three or four
feet wide—covered with bottles, some standing upright, while others
were carefully packed away. There was one shelf where the bottles
had been lying undisturbed for twenty years; another where they
had not been touched for thirty, another for forty, and another for
fifty years. Above most of the shelves a date was chiselled into the
rock, and the date, as I was told, indicated the time when the wine
was bottled and placed there. These chiselled places were, however,
comparatively few, as the most common designation was that of a
date cut in a small piece of board which rested above the bottles.
In some places the dust of ages had
almost obliterated the dates, but our guide
seemed to know them all from recollection. I remember one date of
1750, another of 1634, and I believe there was one board dated
somewhere about 1590. Shelves were pointed out which were said
to contain wine that had not been moved or disturbed in any way for
three hundred years. I do not vouch for the truth of the statement,
but merely give it as I heard it.
It was interesting to observe how the dust and cobwebs had
gathered about the bottles, and also to observe the shapes of the
bottles. The more recent shapes were those familiar to all drinkers
and friends of drinkers of the present day. Then there were short,
thick-set bottles, while others were dumpy and very long in the
neck, reminding one of an overfed goose or a camel suffering with
the dropsy. Some of the earlier bottles indicated that the art of
blowing glass was not well known at the time of their construction,
as they were badly shaped, and frequently had deep indentations in
their sides. Some of them could be called flasks, rather than bottles,
as they had no necks at all, and were round at both ends. All the

ONE HUNDRED DOLLARS
A BOTTLE.
bottles that I examined were carefully sealed, and I was shown
several bottles with long, tapering necks, that had been tightly
closed by melting their ends in a flame after the wine had been
placed inside, just as the tube of a thermometer is closed after it has
been filled with quicksilver or alcohol. In order to get at the wine
enclosed in this way, it is necessary to break away the top of the
neck.
The cellar was perfectly dry, so that no
moisture collected anywhere. I may
remark, by the way, that a dry cellar is
always desirable. There was no moisture, but there was a liberal
supply of dust and cobwebs. On bottles that had been in their places
only a few years, there would be a slight film or covering of dust.
Those that could boast of twenty years, and those that had
remained undisturbed a hundred or two hundred years, were
covered so thickly that it was almost impossible to distinguish the
bottles from the mass which covered them. I saw one shelf—I forget
its age—where not a bottle was visible; it seemed to be a mass of
cobwebs, and nothing more. To judge from its appearance, I would
not have given twenty-five cents for the contents of that shelf; but if
I had offered twenty-five hundred dollars, my offer would have been
spurned with disdain. I asked the value of the wine on this shelf, and
was told that it was twenty guineas a bottle. I did not want any of it
at that price, but I presume that there are plenty of men in the
world who are ready to pay it.
After we had seen the curiosities of the place, the proprietor
insisted that we should make a practical test of his wine. He did not
open any of the twenty-guinea stuff, and we could not expect him
to, though I secretly hoped he would consider himself sufficiently
honored by our presence to do the handsome thing, and break a
bottle or two of it just to give us a taste. The best he would do was
to open a ten-guinea bottle from another shelf. It is not every day
you can smack your lips over wine worth fifty dollars in gold a bottle,
and we sipped it very carefully, and allowed it to trickle not too

WINE TASTING AND ITS
EFFECT.
rapidly down our throats. I found it a very agreeable wine; it had a
rich and fruity, though rather sweetish taste. I know nothing to
which it can be compared, and therefore I will not make any
comparison.
The proprietor treated us on the
descending scale, for the next bottle he
brought us was a five-guinea one. It was
only forty or fifty years old, a very juvenile stuff, but we were unable
to discover any great difference between it and the other. Two or
three kinds of this wine were shown us, and then he brought all
sorts of new wines just in the cellar, that is to say, they had only
been there some five or ten, or it may be twenty years. Other wines
were brought forward for our deglutition; and after a time the thing
became a little monotonous, and I suspected that we might get our
heads and feet a little tangled. I suggested that we had other
business to attend to, and had better not indulge in the wine
business any longer; but the proprietor was polite, and was
constantly offering us just one more sample.
“Have the gentlemen taste this one,” he would say to the officer
who accompanied us, and at the urgent request of the officer we
would indulge the proprietor.
The officer repeatedly stated, on presenting the wine, that that
would be the last; but somehow there was always something new to
be tasted, and something that we could not decline without giving
offence. Before we got through, we tasted nearly every wine in the
cellar, and finally asked to be let off.
When we reached the foot of the stairway, we found it had
shrunken greatly in size. We had descended without difficulty, but
now it was necessary to move up edgewise, and I firmly believe,
that if we had remained below much longer, the shrinking process
would have made the staircase so narrow, and the roof above so
low, that we should have been unable to get out, and might have

TURNING AN AMERICAN
HEAD.
staid there forever. Think of one’s terrible fate in being shut up in a
wine cellar to die.
My companion wanted to sit down on
the foot of the stairs and go to sleep, but I
told him it was not a custom in Poland on
visiting wine cellars, or, so far as I knew, in any other country. He
then asked me to write to his friends, if I succeeded in getting out,
and tell them to send money enough to buy out the concern to take
it home to America. He would take cellar and all if he had to carry
the whole city of Warsaw and the Ex-King of Poland in his trunk. He
had a friend at New York who would just like this sort of thing. He
would be willing to sell all his interest in the United States if he could
only assemble his friends in that cellar, and get them as blind drunk
as he was. I saw that he was wandering mentally, although unable
to wander much physically, owing to the extreme suppleness of his
legs. He began to chide me for taking so much wine, and said I
ought to have followed his example, and drank nothing.
The situation became alarming. There was the staircase growing
narrower until it resembled a loophole in the wall of a fortress. I was
very much inclined to sit down with my friend, and wait until the
place grew larger. While thinking what to do, we were roused by the
appeal of our officer comrade to taste of another wine, a very
superior article from Hungary. We told him politely that we must
refuse, intimated that we should feel much better without it, and if
he could only plan some way by which we could get out of that
cellar and reach our hotel, we should be very much obliged.
He led the way up stairs. We observed that luckily they were large
enough for him to ascend without difficulty, and finally we reached
the space above. Once there we breathed more easily. We thanked
our host for the attention he had shown us; we thanked him by
shaking his hand, and keeping our mouths closed. To thank him in
English would do no good, as he did not understand our language,
and we were a little doubtful of our ability to pronounce our words
correctly. I am sorry that my friend made so free with this ancient

A DANGEROUS BRIDGE.
wine, as it totally incapacitated him from saying a word in Polish or
any other language with which he was not familiar.
When we reached the open air we found that our heads became
level again, and in a little while the effect of our wine-sampling
excursion had passed away. Assuming the dignity of a couple of
emperors, we rode to our hotel, took a lunch, and felt better.
All over the world it is a trick of the proprietors of wine cellars to
put their visitors through the system of sampling, so that, drink as
sparingly as they may,—a teaspoonful at a time only,—they will be
very much confused in body and mind before they emerge from the
clutches of their entertainers.
In one of the Western States I am
acquainted with a wine dealer whose cellar
is entered by crossing a narrow bridge over a brook. The bridge is
ten or twelve feet long, about three feet wide, and has no railing. I
have heard him say that no visitor to his wine vaults ever yet walked
that plank on his return from the cellar without tumbling into the
brook. From what I have heard of his establishment, I think he is not
very far from the truth. Many a visitor to that cellar has received an
involuntary plunge bath as he came out into the open air.

XXIV.
THE BASTILLE.
ITS HISTORY AND CONSTRUCTION.—THREE AMERICANS SEARCHING FOR IT.—A
FRENCH JOKE AT THEIR EXPENSE.—HOW PRISONERS WERE RECEIVED AND
TREATED.—HORRIBLE DUNGEONS.—THE OUBLIETTES.—CRUELTIES OF THE
BASTILLE.—THE MAN IN THE IRON MASK.—HIS ROMANTIC STORY.—
DESTRUCTION OF THE BASTILLE.
One of the most famous dungeons or prisons in the world was the
Bastille of Paris.
It was a state prison and citadel of the city, was built in the year
1369, and destroyed by the mob in the beginning of the revolution
of 1789, or more than four centuries after its construction.
It is a curious fact that no plan of the Bastille as originally
constructed is in existence, neither is there any plan extant of the
Bastille as it appeared at the time of its destruction. Somehow the
kings of France were averse to giving the public much information
about this famous prison of state. They appear to have been
satisfied with the knowledge that the place existed, and that those
who displeased them could be shut up there, and they never
troubled themselves to know the exact plan or model of the concern.
There has been a great deal of exaggeration concerning the
Bastille, and many stories have been told about it which had little or
no foundation. After all, there was really no need of exaggeration,
for the atrocities committed within the walls of the Bastille are quite
horrible enough for all practical purposes.

THE GRAND HOTEL, PARIS.
In ordinary life the French are a quiet, harmless people, and they
are the last in the world whom you would suspect of atrocities; but
every revolution in France has been full of horror, whether in past
times or in the present. It has been said that you may take the
mildest Frenchman in the world, give him a place of authority where
his acts will not be called into question, and the chances are great
that he will conduct himself in a very savage manner. I do not assert
this of my own knowledge, but leave the reader to judge whether
the history of the French prisons and French tyranny does not, in
some degree at least, corroborate the statement.
The day after my arrival in Paris, a friend proposed that we should
visit the Bastille. We were talking upon some topic, and I had
actually stepped inside the carriage with him and given the order to
the driver before it occurred to me that the Bastille did not exist, and
had not existed for several scores of years. When I remembered
this, and told my companion, he said,—

SEARCHING FOR THE
BASTILLE.
VERDANT AMERICANS.
“I came very near selling you. I want to get even on selling
myself.”
Then he told me a story of his
experience in searching for the Bastille.
Bear in mind that he was an editor,
familiar with history (editors of course know everything), and if he
had given the subject a moment’s thought it would have occurred to
him that there was no Bastille in Paris worth mentioning. Let me tell
his story as he told it.
“There were three of us who came over in the steamer, landed at
Brest, and came to Paris. We arrived here in the evening. We put up
at the Grand Hotel, and the next morning started out to ‘do’ the city.
The first thing we saw as we stepped out of the hotel door to the
Boulevard was an omnibus, on which was the sign ‘Place de la
Bastille.’ We mounted to the top of this omnibus, and away we rode
down the Boulevard.
“By and by we stopped near a large,
open square, with a monument in the
centre. The conductor motioned us to get off, and said something
which we did not understand, but took to mean that this was the
end of his route. Moreover, the omnibus turned round, and we
understood pretty well that we must get ashore. I was the only one
who could speak French, and I couldn’t speak much of it. As we left
the omnibus, I said to the conductor, ‘Monsieur, où est la Bastille?’
“The conductor stared at us, smiled, and turned away. Then we
stepped on the sidewalk and looked around. Close by us was a
‘Restaurant de la Bastille,’ and on the corner we could see the sign of
‘Place de la Bastille.’ There was a cake shop close by, and that had a
sign which indicated that it was the cake shop ‘de la Bastille.’
“Then we stopped a well-dressed Frenchman, and said to him,
‘Monsieur, où est la Bastille?’ The fellow was too polite to laugh in
our faces, as the conductor did, but he said not a word, and walked

DESCRIPTION OF THE
BASTILLE.
off. I saw, though, when his back was turned towards us, that he
was shaking his sides, and evidently grinning.
“Then we stepped into the restaurant, and I said to a waiter,
‘Garçon, où est la Bastille?’ and that infernal waiter laughed in my
face. I said to the other boys, ‘These confounded Frenchmen round
the Bastille are all fools. I thought Frenchmen were polite, but these
fellows have no politeness at all.’ We climbed out of that restaurant,
and went out on the square on a Bastille hunt.
“There was no more sign of a prison than there is inside your
boot. We walked round that square about ten minutes, when it got
into one of our heads,—not into mine though,—that the Bastille had
been destroyed in 1789. I had nothing more to say, except that we
were the three biggest fools in all Paris. Here we had been hunting
round, boring everybody, and asking them to show us a prison which
was destroyed eighty years before, as we perfectly well knew, only
we did not happen to recollect it. We went back to the Grand Hotel,
and the next time we went out sight-seeing we made sure that the
thing we inquired for was in existence.”
The Bastille was an irregular building in
shape, as the original construction, in the
time of Charles V. had been added to by
each successive monarch. It had as its principal feature eight round
towers, connected by curtains of masonry, and was encircled by a
ditch a hundred and twenty-five feet wide. This ditch was generally
dry, and was surrounded on its outside by a wall sixty feet high, to
which was attached a wooden gallery running round the whole inner
circumference of the ditch opposite the castle. This gallery was
called the “Rounds.” Sentinels were stationed on these Rounds, and
it was their duty to be perpetually in motion, in order to discover any
movement of the prisoners for escaping. The Bastille had a governor
and a staff of assistants, and it had a garrison of one hundred men,
with their proper officers.

CHARACTER OF THE
DUNGEONS.
Whenever a prisoner was brought to the Bastille, his trunks and
clothing were carefully examined, in order to discover whether he
had any concealed papers or weapons. The advocate Linguet, who
had been detained there for three years, says,—
“The new comer is as much surprised as alarmed to find himself
subjected to a personal examination by four men, whose appearance
seems to belie their functions; men clad in uniforms, which leads
one to look for a regard to decencies, and wearing decorations
which presuppose a service which endures no stain. This man takes
from him his money, that he may have no means of corrupting any
one of their number, his jewelry on the same consideration, his
papers for fear he should find any resource against the tedium to
which he is henceforth devoted, and his knives and scissors are
taken from him for fear he should commit suicide or assassinate his
jailers.”
After this examination he was led to the cell intended for him to
occupy. These cells were situated in all the towers. The walls were at
least twelve feet in thickness at the top, and at the base they were
thirty or forty feet. Each cell had a small window defended by three
iron gratings, one within, the second without, and the third in the
middle thickness of the masonry.
The bars of this grating were an inch thick. No fire was allowed,
and there was no glass in the windows, so that in winter these cells
were like ice-houses, and in summer they were hot and damp.
The dungeons were nineteen feet below
the level of the court-yard, and five below
that of the ditch. They had no openings
but a narrow loophole communicating with the ditch. The inhabitant
of these dungeons was deprived of air and daylight, and lived in a
damp and infected atmosphere. Oftentimes the floor of his cell was
covered with mud, and he found himself surrounded by reptiles,
rats, and other disagreeable creeping or walking things.

The written history of the Bastille shows that these horrible cells
were frequently used for the confinement of prisoners in order to
make their existence as terrible as possible. There is a tradition that
iron cages were used for the confinement of prisoners, but writers
who have given their attention to this subject say that nothing of the
sort was discovered at the time the Bastille was destroyed. There is
also a tradition in regard to the Oubliettes, which are described as
holes into which condemned prisoners were lowered, where they
should languish and die forgotten. There is also a tradition in regard
to a Question Chamber, in which suspected prisoners were tortured
to make them confess their guilt, or to reveal the names of their
accomplices.

PLACE DE LA BASTILLE, PARIS.
THE BASTILLE.—ERECTED IN 1369.
The Bastille could contain fifty state prisoners in solitary cells, and
by putting two persons in one cell the number could be raised to a
hundred. Sometimes as many as three hundred persons were in the
Bastille at once, and in that case they were densely crowded.
According to history the prisoners were wretchedly fed, but it should
be said, in justice to the government, that this state of affairs was
probably due to the frauds of the subordinates rather than to any
intended cruelty on the part of the government, as the latter
generally made liberal allowances for the support of the prisoners of
state. One writer asserts that in his time the governor of the Bastille
had a great number of prisoners, many of whom were paid for at
twenty-five francs a day, and that their subsistence did not cost as
many sous. There was a regular tariff for expenses for the table,

TREATMENT OF
PRISONERS.
lights, and washing of all prisoners, according to their rank. A prince
was allowed fifty francs a day, a marshal of France thirty-six francs,
a lieutenant general thirty-four francs, and so on down to the inferior
prisoners, who were allowed two francs and a half.
A prisoner might be examined at the
moment of his arrest, or not until weeks,
months, or years afterwards. He had no
mode of offering any defence, or of telling his friends where he was,
or why he was detained; and sometimes he did not himself know
these facts. He was allowed no books or papers; he could not
communicate with anybody except by special permission. He could
not be visited except on an order from the lieutenant of police, and
at such visits all the conversation must be in the presence of an
officer of the prison, and no allusion could be made to the cause of
detention, the term of imprisonment, or any topic of that sort.
The treatment of prisoners varied greatly. Some, whom it was
desired to kill by slow torture, without trial, or even without a
hearing, were shut up in the horrible dungeons already described,
where they were fed on the worst possible food until death relieved
them from their suffering. Others, whom it was not designed to
punish or destroy, but simply to detain, enjoyed every comfort, and
a great deal of luxury. They had large rooms, fine furniture, excellent
and abundant food, plenty of wine, books, and papers, could have
their own servants, could be visited by their friends or families; in
fact, could do pretty nearly as they pleased, except to go out of the
Bastille.
Sometimes the Bastille was under governors who had a good deal
of the milk of human kindness in their composition, and sometimes it
was under the control of men who had as little feeling and sympathy
as a stone. Prisoners were well or badly treated according as the
governor was good or bad in character, and also according to the
instructions which had been received concerning their treatment.
The most horrible feature about the Bastille was the mode of
sending persons to it. No man could be safe from imprisonment

Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade
Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.
Let us accompany you on the journey of exploring knowledge and
personal growth!
ebookultra.com

Download ebooks file Mediating between Concepts and Grammar Holden Härtl (Editor) all chapters

More Related Content

Similar to Download ebooks file Mediating between Concepts and Grammar Holden Härtl (Editor) all chapters (20)

Recently uploaded (20)

Download ebooks file Mediating between Concepts and Grammar Holden Härtl (Editor) all chapters