LINGUIST List 14.224

Tue Jan 21 2003

Review: Pragmatics: Wallis, Otheguy, Stern (eds) (2002)

Editor for this issue: Naomi Ogasawara <>

What follows is a review or discussion note contributed to our Book Discussion Forum. We expect discussions to be informal and interactive; and the author of the book discussed is cordially invited to join in. If you are interested in leading a book discussion, look for books announced on LINGUIST as "available for review." Then contact Simin Karimi at


  1. Giampaolo Poletto, Wallis, Otheguy, Stern (eds) (2002), Signal, Meaning, and Message

Message 1: Wallis, Otheguy, Stern (eds) (2002), Signal, Meaning, and Message

Date: Tue, 21 Jan 2003 15:01:34 +0000
From: Giampaolo Poletto <>
Subject: Wallis, Otheguy, Stern (eds) (2002), Signal, Meaning, and Message

Wallis, Reid; Otheguy, Ricardo; Stern, Nancy eds. (2002) Signal,
Meaning and Message. Perspectives on sign-based linguistics. John
Benjamins Publishing Company. Amsterdam/Philadelphia. SFSL 48. ISBN 90
272 1557 X (Eur.) - 1 5811 289 6 (US).

Book Announcement on Linguist:

Giampaolo Poletto,
University of P�cs (HU), 2nd year Applied Linguistics PhD student.

The comprehensive discussion of Columbia School Linguistics
conferences, where the sign-based theory was exposed (see
Contini-Morava and Goldberg 1995), lays the premises of this second
volume, where a collection of sign-based inspired studies addresses
specialists, aiming to frame the contribution to linguistics. The
book is in three parts, bibliographical references are provided at the
end of each paper, with tables, eventual appendixes and notes, indexes
of names and subjects follow in conclusion. Contents cover two
domains, familiarity with Columbia School sign-based theory and
elaboration of themes connected to the papers, as specified in the
introduction, where the conceptual axes along which the work unfolds
are clearly defined, together with a brief on the papers below, not in
the order they are presented in the volume.

Part I Theoretical and Methodological Issues

(What) do noun class markers mean?
Ellen Contini-Morava, 3-64.

Sign-oriented and cognitive-oriented approaches (see Langacker 1987,
Reid 1991) to linguistic structures are integrated in the analysis of
noun class markers in Swahili (see Dixon 1986, Contini-Morava
1996). Far more numerous than European gender classes, and each with
internal semantic structure, Swahili noun class markers share no
single set of defining semantic characteristics with associated
classes of nouns. The metaphoric, metonymic and associative
connections nouns form link them into a semantic network. Switching a
marker may thus alterate the interpretation of a stem. Markers are
impervious to analyses as straightforward meaning-bearing units,
because the variable or intermittent effect does not allow to yield a
consistent semantic value. They have meaning but class members have
nothing in common. By analysing how three Swahili noun classes are
structured semantically the cognitive perspective of the semantically
related senses of a lexical item is envisioned. By focusing on their
communicative function, they are discovered to maximize the use of
pronouns instead of nouns. In the end, one out the four analytical
options provided for the markers is chosen, because the semantic
diversity of each class is recognized and each marker is posited a
single value. 

Rethinking the Place of Statistics in Columbia School Analysis. 
Joseph Davis, 65-90. 

As to the role of statistics, Davis rejects chi-square test, often
inappropriately used, due to the lack of statistical independence for
observations in connected texts (see Woods et al. 1986). Inferences
are not reliable because no ideal langue, no linguistic community, no
population to randomically draw tokens from are
envisioned. Furthermore, observations to be accounted for do not need
to claim great generality, and the prevalence of a linguistic feature
is of little or no interest in such studies. After providing tables,
references and examples (see Davis 1992), the author concludes the
role of statistics in Columbia School grammatical analysis to be
relatively small, and occurring near the end of a definite modus
operandi suitably supportive of a hypothesis, which may as well
include analyzing a variety of genres or using inferential statistics.

The Linguistic Sign in Its Paradigmatic Context: Autonomy Revisited.
Mark J.Elson, 91-110. 

>From Bybee and Brewer's hypothesis (see 1980), and with other
sign-based linguists' premise that the distribution of a form is a
function of the meaning it bears, the issue of morphological change in
verbal paradigm is here seen as distributional. The focus is on the
autonomous feature of certain forms of verbal paradigms, supposed to
serve as base to generate other forms, especially on the third, first
and second person singular, in a scale of autonomy. Semantic
complexity, opacity and frequency of occurrence are the factors the
above hierarchy is based on. In a sign-based perspective, a member of
verbal paradigm may be under-specified grammatically, given some
circumstances; its expressive potential is superior to that of more
specified forms, which makes it suitable as a word-level base for
formal innovation, as with a dubbed 'first person singular' compatible
with a first person plural.
Part II Sign-Based Linguistic Analyses 

A Surpassingly Simple Analysis. 
Joseph Davis, 113-136.

Skepticims about traditional categories and the notion of human
language complexity (see Pinker 1994) nurture the author's study on
pronominal reference in Italian, as to the third-person pronouns ess+,
loro and s�, disjunctive demonstrative, reciprocal and reflexive
respectively, characteristics reflected in grammatical categories
which are found not matching the forms' use in texts, where they are
all employed for both reflexive and non-reflexive reference. Davis
then resorts to and tests the notion of information load: each pronoun
signals an amount of - gender and number's information on its
referent. Pronouns are therefore chosen according to what the reader
needs to find their referent in the context.

Serb-Croatian Deixis: Balancing Attention with Difficulty in Processing. 
Radmila J.Gorup, 137-156.

The 2-tiered explanatory role of meaning underlies the analytical
focus Radmila Gorup poses on the grammar-morphological issue of the
3-member Deixis (D) system of Serb-Croatian pronouns. They correspond
to the English 'this' and 'that', and indicate a High, Mid and Low
(HD, MD, LD) level of attention, depending on the pronoun referent
appearance in discourse, as a speaker's specific signal (see Garcia
1975). It is part of a referent-finding strategy helping the hearer to
locate the referent, with the formal-semantic correlate: HD,
hard-to-find and difficult to process; MD, of medium difficulty and
recently mentioned through a Noun; LD, with little or no difficulty
and no particular communicative referent. Data are then statistically
tested through contexts with deictic pronouns. 

Do's One Sign, One Meaning? 
Walter Hirtle, 157-170.

Hirtle provides a further example of setting aside traditional
distinctions, here syntactic, after Ruhl, lexical, and Davis,
grammatical, through the analysis of the English do, in the search for
semantic unity in its auxiliary, suppletive and main verb capacity. A
progression of semantic specification is detected, as to duration,
duration and change, duration, change and transitivity
respectively. To capture this and maintain the sign unity, the author
makes use of the concept of ideogenesis (see Guillaume 1984, 1987). A
potential meaning links to operative conditions realizing portions of
it in discourse, and turns out to delineate a semantic trajectory from
minimal to maximal specification, along which the reader moves and
stops, when the point for the degree of actualization of his mentally
represented experience is found. Ideogenesis hence includes the word
actual meanings observable in discourse. 

Data, Comprehensiveness, Monosemy. 
Charles Ruhl, 171-190.

Aprioristic notions are here set aside in the lexical meaning area of
a dictionary word multiple definitions, mainly borrowed from classical
rethoric and philosophy. Wha alternatively emerges is the
Comprehensiveness Principle, according to which the measure of a
word's semantic contribution is not accuracy (in a single context) but
comprehensiveness (in all contexts). The sign-based principle and
dictionary tradition confrontation parallels monosemic- and
polysemic-oriented research (see Fillmore 1970, Bolinger 1971, Ruhl
1989), with the latter definitely more plausible with a restricted
data-base. When augmented with examples from language use, namely
forty-nine, an interpretive continuum is created between the chosen
items, breaking sticks and breaking in a new man, along with the idea
of two discrete ''senses'' of break, whose following monosemic
analysis, supported by further eighty-three examples, prove how
information comes from the surrounding context and has been
misattributed by the polysemic approach. 

Phonology As Human Behaviour: Initial Consonant Clusters Across Languages.
Yishai Tobin, 191-256.

With a thorough theoretical introduction, by means of detailed
documentary tables (see Diver 1979, Davis 1984/1987, among the others)
to support data provided for a variety of languages, Tobin deals with
world languages as to some word-initial consonant clusters,
classifiable as more frequent, less frequent or non-existent. One
syllable words in English, for instance, have a frequent tr / sl
combination of initial consonants, a rare - three cases - sf, a
non-existent tl / sr. Such synchronic language facts are assumed to be
due to diachronic pressures operating on speakers' selection of
lexical items. Pressure originates from the difficulty of articulating
either individual phonemes or sequences of phonemes in a
language. Principles act over time up to a morphological natural
selection which eliminates functionally less advantageous specimens,
namely hard-to-pronounce words, excluded when lexical alternative
exists. In phonological evolution there is an intrinsic articulatory
advantage, lying on three principles: similar articulatory gestures
are easier to pronounce; additional articulators increase pronounce
difficulty; same articulator reuse in near phonetic environments is
difficult. Predictions consequently to be tested on any language lay
on data from these languages: thirty Indo-European, three Semitic,
three Ugro-Finnic and one Caucasian. Sound changes occurred before
completion of the evolutionary process may result in failures. 

Celtic Sense in Saxon Garb. 
Michael P. Wherrity, 257-272.

Another cognitive inspired treatment is displayed in a study of Irish
influence on American usage of a frequently occurring on + personal
pronoun or noun structure, for an event happening to the disadvantage
of the party involved. This parallels Irish Gaelic usage, but neither
loan translation nor calque are suitable explanations to something
structurally innovative in the English of Irish Americans. There is no
semantic change in usage, either. An alternative analysis is offered,
then, following Otheguy's (see Otheguy 1995). Innovative for English
speakers and commonplace for Irish Gaelic speakers, structure and
meaning originate out of 'conceptual affinity', as an extension of
already established on messages, such as to play a joke on someone. If
English was systemically capable of expressing the Irish-inspired
message, speakers were not until exposed to the Irish model. 

Problems of Aspirations in Modern Standard Urdu. 
Abdul Azim, 273-308.

Aspiration in Urdu, rich in stop consonants, allows the author to
ideally extend Tobin's functional phonology. The p - t - k phonemes,
together with the retroflex t., a palatal c, a post-dursum q, have all
but the last aspirated versions (ph - th - t.h - ch - kh) and voiced
versions (b - d - d. - j - g), which on their turn have an aspirated
version (bh - dh - d.h - jh - gh), a unique feature in Indic families.
The four-way classification - in the paper fully displayed - differs
as both to their positions within the morpheme and the frequency in
the lexicon, accordingly to their articulatory complexity (see Perkins
and Kent 1986) and the functional loads at the word beginning or
end. At the level of a systemic and typological asymmetry of phonemic
inventory, two anomalous facts manifest: no garden variety h phoneme,
despite using aspiration as a 'complicating factor' in two sets; a
typologically rare and articulatorally unnatural voiced h. Due to the
ease of acquisition factor, Azim maintains, a phoneme whose
acquisition facilitates other phonemes's acquisition rightly has an
ecological niche; the complex reasons for simple h absence in Urdu are
then given. 

Part III Columbia School in the Context of 20th Century Linguistics

Cognitive and Semiotic Modes of Explanation in Functional Grammar.
Alan Huffman, 311-338.

Cognitive linguistics and sign-based theory modes of explanation find
a synthesis by confronting the views on word order, whose use is
''iconic'' and ''natural'' or which is the signal of a meaning
respectively. The full-verb inversion in English is not tackled in
non-semantic, syntactic terms, consistent with paradigms in
traditional linguistic analyses, treating the subject-verb order to
attribute post-posed subjects to syntactic triggers. S-V and V-S are
sequences to be viewed as signals of grammatical meanings relative to
the discourse focus (see Birner 1992). The use of each hints at the
speaker's desire precisely to signal the relevant meaning;
consequently the desire hints at the meaning discourse function in the
narrative structuring. That demonstrate the quantitative and
qualitative analysis of 'Pioneers!' and 'Lord of the Flies'. 

The Future of a Minimalist Linguistics in a Maximalist World. 
Robert S.Kirsner, 339-372. 

Major criticisms of Columbia School, especially from cognitive
linguists, are here examined: a too reductionist approach, for
postulating sparse meanings for linguistic forms; unconvincing
analyses, when pragmatic factors explain how they convey concrete
messages; analytical control, not psychologically grounded. A first
response is that meanings are as too sparse in Columbia School as too
complex when precise out of massive polysemy for successful
communication. A second one is that linguistic theories are not
neurological structure, which is proved by other theories similar
shortcomings in psychological guidance. A final one considers language
not a self-contained representational system, rather a communicative
tool knowledge experience and contextual factors enable to
function. In the end, Columbia School can rightly contribute to
linguistics as to its therapeutic function, in forcing others to look
at problems and data from a different perspective, and its empirical
methodology, involving genuinely experimental techniques to be
continuously developed and refined. 

Saussurean Anti-Nomenclaturism in Grammatical Analysis: 
A Comparative Theoretical Perspective. 
Ricardo Otheguy, 373-404. 

Saussurean, Chomskyan, traditional and Columbia School positions are
duly exemplified and confronted (see Chomsky 1957, Harris 1988,
Nichols & Woodbury 1984, Reid 1991). The epistemological grondwork for
the search for objectivity is here comparatively laid starting from
Saussurean radical anti-nomenclaturism as to linguistic categories, to
be discovered by linguistic principles, such as the linguistic
sign. They are posited as not deriving from philosophy or logic,
without the external motivation western tradition has been relying on,
due to the fact that language is a system unto itself, thus discarding
the concepts of meaning as reference and language as cognition, and
that meaning and structure are both language internal and language
particular. Given the signifiant and signifi� fixed association, a
language is to be approached setting aside all traditional grammatical
categories, whereas semantic and structural ones hold no privileged
status, rather develop hypotheses to be tested out, and looking for
categories in regular relation to form. 


The knowledge of the fist volume contents, here constantly recalled,
is necessary to tie together and frame otherwise distant points. There
is a unique methodological empirical perspective and many objects of
investigation to support mainly comparative theoretical approaches. To
embed all in a coherent framework is not an easy task: there are past
linguistic theories, actual studies and the future of Columbia School.
The scenario is diachronically wide and sinchronically complex, as
each paper displaying the analysis of a specific issue shows. Just the
gap from general premises to very particular insights may result hard
to fill sometimes, unless basic notions provided are immediately
clear. Accuracy and documentary thoroughness are on the other hand
remarkable; that testifies - and is mostly to appreciate - how
concreteness is extremely relevant in this functionalist approach,
resulting in analyses which supply with a variety of valuable data,
perspectives and suggestions. 


Birner, Bety j. 1992. The Discourse Function of Inversion in English.
Ph.D. dissertation, Northwestern University.

Bolinger, D. 1971. "Semantic overloading: a restudy of the verb
remind" Language 47: 555-73.

Bybee, J. and M. Brewer. 1980. "Explanation in Morphophonemics:
Changes and Proven�al and Spanish preterite forms." Lingua 52(3/4):

Chomsky, Noam. 1957. Syntactic Structures. The Hague: Mouton
Publishers. Contini-Morava, Ellen and Barbara Sussman Goldsberg
(eds). 1995. Meaning as Explanation: Advances in linguistic sign
theory. Berlin: Mouton de Gruyter.

Contini-Morava, Ellen 1996 "Things - in a Noun Class Language: Semanic
functions of grammatical agreement in Swahili". In Edna Andrews and
Yishai Tobin (eds), Toward a Calculus of Meaning: Studies in
markedness, distinctive features and deixis. Amsterdam and
Philadelphia: Benjamins.

Davis, Joseph. 1984/1987. "A Combinatory Phonology of Italian."
Columbia University Working Papers in Linguistics 8: 1-99.

Davis, Joseph. 1992. Italian egli and lui: Grammatical meaning and
inference. Ph.D. dissertation, Columbia University.

Diver, William. 1979. "Phonology as Human Behavior." In D. Aaronson
and P. Reiber (eds), Psycholinguistic Research: Implications and
applications. Hillside NJ: Lawrence Erlbaum, 161-186.

Dixon, R.M.W. 1986. "Noun classes and Noun Classification in
Typological Perspective." In Colette Craig (ed), Noun classes and
categorization. Amsterdam: Benjamins, 105-112.

Fillmore, C. 1970. "The grammar of hitting and breaking." In
R. Jacobs and P. Rosenbaum (eds), Readings in Transformational
Grammar. Waltham, MA: Ginn, 120-33.

Garcia, Erica. 1975. The Role of Theory in Linguistic Analysis: The
Spanish Pronoun System. Amsterdam: North Holland.
Guillaume, Gustave. 1984. Foundations for a Science of Language.
Amsterdam and Philadelphia: Benjamins.

Guillaume, Gustave. 1987. Le�ons de linguistique de Gustave Guillaume
1947-1948, s�rie C: Grammaire particuli�re du fran�ais et grammaire
g�n�rale III. Quebec: Presses de la Universit� Laval et Lille,
Presses Universitaires de Lille.

Harris, Roy. 198. Language, Saussure, and Wittgenstein: How to pla
games with words. London and New ork: Routledge.

Langacker, Ronald. 1987. Foundations of Cognitive Grammar. Stanford,
CA: Stanford University Press.

Nichols, Johanna and Anthony Woodbury (eds). 1984. Grammar Inside and
Outside the Clause: Some views of theory from the field. Cambridge
University Press.

Otheguy, R. 1995. "When Contact Speakers Talk, Linguistic Theory
Listens." In E. Contini-Morava and B. Sussman Goldberg (eds), Meaning
as Explanation: Advances in linguistic sign theory. Berlin: Mouton de

Perkins, William H. and Raymond D. Kent. 1986. Functional Anatomy of
Speech, Language, and Hearing. San Diego, CA: College-Hill Press.

Pinker, Steven. 1994. The Language Istinct. New York: Harper Perennial.

Reid, Wallis. 1991. Verb and Noun Number in English: A functional
explanation. London and New ork: Longman Publishers.

Ruhl, C. 1989. On Monosemy. Albany, NY: SUNY Press. 

Woods, Anthony, Paul Fletcher, and Arthur Hughes. 1986. Statistics in
Language Studies. Cambridge University Press. 1993.


Bachelor in Foreign Languages and Literature, English and Russian, and
Humanities in Italy, Giampaolo Poletto is a second year PhD student in
Applied Linguistics at the University of P�cs, in Hungary, with
teaching qualifications for secondary schools in English and in
Italian, taught in Italy and abroad for ten years, with a research
project which, by focusing on humor, combines linguistics and
teaching, with a pragmatic and semantic analysis of a corpus of texts
and a didactic synthesis through the production of possibly
multimedial material for Italian S/FL students; that should sort of
collect past personal teaching experiences and studies, feed a
linguistic and thematically oriented research programme, open new work
and study perspectives.
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue