27.2749, Review: Ling Theories; Morphology; Typology: Baerman, Brown, Corbett (2015)

[The LINGUIST List via LINGUIST]

LINGUIST List: Vol-27-2749. Mon Jun 27 2016. ISSN: 1069 - 4875.

Subject: 27.2749, Review: Ling Theories; Morphology; Typology: Baerman, Brown, Corbett (2015)

Moderators: linguist at linguistlist.org (Damir Cavar, Malgorzata E. Cavar)
Reviews: reviews at linguistlist.org (Anthony Aristar, Helen Aristar-Dry, Robert Coté, Sara Couture)
Homepage: http://linguistlist.org

*****************    LINGUIST List Support    *****************
                       Fund Drive 2016
                   25 years of LINGUIST List!
Please support the LL editors and operation with a donation at:
           http://funddrive.linguistlist.org/donate/

Editor for this issue: Michael Czerniakowski <mike at linguistlist.org>
================================================================


Date: Mon, 27 Jun 2016 13:43:32
From: Anish Koshy [anish at efluniversity.ac.in]
Subject: Understanding and Measuring Morphological Complexity

 
Discuss this message:
http://linguistlist.org/pubs/reviews/get-review.cfm?subid=36145917


Book announced at http://linguistlist.org/issues/26/26-1909.html

EDITOR: Matthew  Baerman
EDITOR: Dunstan  Brown
EDITOR: Greville G. Corbett
TITLE: Understanding and Measuring Morphological Complexity
PUBLISHER: Oxford University Press
YEAR: 2015

REVIEWER: Anish Koshy, English and Foreign Languages University, Hyderabad

Reviews Editor: Helen Aristar Dry

INTRODUCTION

Morphological complexity can be a result of various factors of exponence, that
is, how individual items are realized. Complexity could be a result of
discontinuous roots, circumfixes, portmanteau morphs, morphological zeroes,
subtractive morphology, umlaut, ablaut, syncretism, defective paradigms,
deponency, variation, allomorphy, historical change, restructuring or
grammaticalization among others, or the result of the sheer quantity of
morphological elaboration of the polysynthetic type. “Understanding and
Measuring Morphological Complexity”, edited by Matthew Baerman, Dunstan Brown,
and Greville G. Corbett, is an important contribution to a very important area
of inquiry in linguistics and is the second one on complexity to be published
recently by the OUP, with the former (Newmeyer and Preston (2014)) dealing
with syntactic complexity from a Generative perspective. This volume consists
of ten chapters arranged in 3 parts with an 11-page common reference-list at
the end.

SUMMARY

In the introductory chapter “Understanding and measuring morphological
complexity: An introduction”, the editors discuss various approaches to
understanding morphological complexity from previous works on the theme. These
approaches distinguish complexity in morphology as not being linked to syntax
and as being results of looking at morphological operations variously from a
syntagmatic or a paradigmatic approach. Entropical analysis of complexity
(based on predictability) by different scholars is also touched upon. The
different papers in the volume are also summarized.

In Chapter 2, “Dimensions of morphological complexity”, Stephen R. Anderson
invokes Sapir’s three-dimensional understanding of morphological markers from
a typological perspective, namely, (a) the number of inflectional and
derivational markers themselves; (b) the way the markers attach (inflectional
vs. agglutinative; the scope that later affixes have on already attached
ones); and (c) the number of markers that can be found on a word (isolating
vs. polysynthetic). Complexity is understood as not necessarily always a
consequence of a system’s components being added up together, but also arising
due to morphotactic organization, non-resolution of the competing demands of
Faithfulness and Markedness, semantic factors of scope, and/or phonological
factors.

Part II on “Understanding Complexity” begins with Chapter 3, “Rhizomorphomes,
meromorphomes, and metamorphomes”, wherein Erich R. Round distinguishes what
he calls three species of morphomes. Morphomes, though prominent in the
organization of a language’s morphological system, are observed to be mostly
anisomorphic with the syntactic, semantic and phonological systems.
Rhizomorphomes deal with classes of roots, which determine their morphological
behaviour; meromorphomes deal with exponence of morphological operations;
metamorphomes deal with realization of meromorphomes. The data discussed in
the chapter is mostly from Kayardild. It is argued that morphological
exponence due to percolating of features and non-realization of some features
due to blocking leads to complexity. Contexts of deponency and virtual rules
of exponence are also posited to account for the system’s complexity.
Kayardild complexity is said to arise due to the ability of ‘identity of
exponence’ to skip some eligible classes of stems and also by their ability to
breach the inflection-derivation divide.

In Chapter 4, “Morphological opacity: Rules of referral in Kanum verbs”, Mark
Donohue focuses on complexity that arises as a result of lack of transparency
in the mapping of features present and their expression. Verbal inflection in
Kanum is an elaborate system of oppositions with mostly regular distribution.
It also shows instances where elements of a paradigm are marked by referral
from other inflectional cells, with pronominal object agreement markers
showing syncretism. This happens in spite of there being independent pronouns
and independent agreement indexing for all persons in the subject role in the
language. Complexity in Kanum is also attributable to rarity (as compared to
other languages), due to the invoking of unusual features for the person axis
in the agreement system. Opacity in the pronominal system also adds to the
complexity; so also does the unpredictability of takeovers of inflectional
cells.

In Chapter 5 “Morphological complexity à la Oneida”, Jean-Pierre Koenig and
Karin Michelson discuss morphological complexity arising out of a system of
morphological referencing of event participants, achieved via a system of
fifty-eight pronominal prefixes, an obligatory part of the verbal inflection.
Complexity arises here out of the paradigmatic notions of selection and
segmentation. Paradigmatic complexity is quantified via the number and kind of
rules needed for realizing morphological distinctions, the presence of
semantic ambiguity, directness/indirectness of rule application, and the ease
of segmentation and generalization. The authors also wonder why the task of
retrieving multiple pronominal prefixes is generally considered more complex
than the task of retrieving content words from the mental lexicon, when the
number of vocabulary stems known to us is in many magnitudes higher. It is
speculated that this could be so because of (a) obligatory nature of
inflectional markers, (b) the inter-related choices speakers have to make, and
(c) the choice of the right allomorph.

In Chapter 6 “Gender–number marking in Archi: Small is complex”, Marina
Chumakina and Greville G. Corbett discuss how gender-number marking, though
realized by a small paradigm of markers, proves complex due to multiple
overlapping factors – phonological, morphological, syntactic and semantic.
Phonological factors include considerations like consonant or vowel initial,
monosyllabic or polysyllabic structure, and stress on first syllable or second
syllable. Morphological factors contributing to complexity include the
presence of aspectual and modality markers, inventory of inflectional targets,
position of the inflectional marker, and patterns of syncretism (with only
suffixes allowing syncretism), among others. Syntactic factors like part of
speech also affect choices. Semantic factors like stative vs. dynamic verbs
and lack of semantic homogeneity also contribute to complexity.

Part III on “Measuring Complexity” begins with Chapter 7 “Contrasting modes of
representation for inflectional systems: Some implications for computing
morphological complexity”. The authors, Gregory Stump and Raphael A. Finkel,
begin with a critique of information-theoretic and set theoretic based
computational models of computing complexity. According to them the
computational models are limited by the manner in which information-class
systems are represented. They discuss the ‘plat’ representation of English
verbs and the series of choices it entails in terms of morphosyntactic
property sets, representation of inflectional exponents, etc. The ‘plat’
representations are discussed from a ‘hearer-oriented’ and a
‘speaker-oriented’ perspective, and also in terms of conditional entropy,
principal parts, inflection-class and cell predictability. According to the
authors complexity of an inflectional system is taken to be the extent to
which it inhibits motivated inferences about the word forms in a lexeme’s
realized paradigm.

In Chapter 8 “Computational complexity of abstractive morphology”, Vito
Pirrelli, Marcello Ferro, and Claudia Marzi discuss the abstractive
perspective where word forms (and not roots and affixes) are treated as basic
units and their recurrent parts as abstractions over full forms. Learning
morphology in this perspective is learning relations between fully stored word
forms. This view is in opposition to the constructive approach, which is
morph-based. The authors contend that the associative framework of the
abstractive approach fares better in explaining morphological processing and
acquisition. Computational complexity is understood by means of
self-organizing maps. The approach presented here represents a
cross-disciplinary approach to language inquiry. It presents a computational
model that tests dynamic inter-relations and processing architectures.

In Chapter 9 “Patterns of syncretism and paradigm complexity: The case of Old
and Middle Indic declension”, Paolo Milizia discusses the principle of
compensation with respect to the massive case syncretism in the dual number of
the nominal declension of ancient Indo-European languages. Syncretism is seen
as a result of a superordinate category, which appears with a marked value.
The pattern becomes complex when the number of grammatical categories involved
is more than two. Nested hierarchies also complicate the idea of a category
hierarchy in these operations. The adjectival declension of the –a/ā –stems in
Old and Middle Indic and the syncretism pattern across three grammatical
categories of number, gender and case are discussed in some detail. The author
argues for a need to reformulate the compensation principle in terms of
frequency and information content, as the principle of compensation basically
achieves avoiding an excessive number of marked values. While marked values
are not addable, markedness is. Paradigm relationships are understood in terms
of information theory. Syncretism is seen as compensating for differences in
frequencies of information cells of a paradigm in order to achieve equilibrium
with respect to the quantitative distribution of inflectional items.

In Chapter 10 “Learning and the complexity of Ø-marking”, Sebastian Bank and
Jochen Trommer discuss the process of ‘stemming’ (identification of lexical
stems in inflectional word-forms) and the process of ‘subanalysis’
(identifying affixes in the affixal strings isolated by stemming). For
example, in German, Ø-affixes are seen as crucial factors enabling
subanalysis. It is seen as facilitating the learning of subsegmentation. A
hierarchy of subanalysis complexity is drawn up based primarily on the
parameters of occurrence of free markers and reduction in search-space for
possible segmentation. An incremental learning algorithm for inflectional
affixes is drawn up to understand its impact on learning of inflectional
paradigms. A correlation between complexity classes and distribution of
Ø-marking is also drawn up based on a typological pilot study of inflectional
verbal paradigms of twenty languages.

EVALUATION

The study of linguistic complexity is riddled with many traditionally
received, axiomatically-accepted-as-true positions about languages. When it
comes to complexity it has often been viewed more like a trivia than a
linguistic tool. The accepted wisdom in the field, also known as ‘compensatory
complexity’ (cf: Bane 2008: 69), is that all languages are equally complex and
that what one language expresses in morphology is equally expressible in
another language by some other means.  This has meant that linguists have
often avoided a comparative study of complexity. Since complexity has often
been dealt with impressionistically, with languages with unfamiliar structures
for the investigating linguist coming out as ‘complex’, such studies have also
been accused of turning languages of unfamiliar cultures into exotic entities.
This volume has to be commended for not only avoiding the pitfalls of the
aforementioned types, but also for attempting to provide numerical and
algorithmic bases for measuring and comparing aspects of complexity, not
merely across languages but also within a language between what may be
impressionistically understood as simple and complex structures. It is only
when we have a substantial literature available on this, that we could move
forward in hypothesizing if morphological complexity, its parameters and its
components constitute in any way a significant typological parameter in the
description of languages comparable to other typological parameters like the
Greenbergian ones.

The concept of complexity in language is not a homogeneous one. Some scholars
have looked at complexity as not representing a language’s morphological
richness but its redundancies and its unproductive patterns (cf: Dressler
(1999)). Others, as in the present volume, look at complexity in morphology as
being directly linked to the complexity of functions and as being about
paradigmatic relationships. Thus, highly agglutinating languages like Turkish
or the Dravidian languages, or inflectional languages like Sanskrit and Latin,
with elaborate paradigms, come out as representing complex morphological
structures. In some of these languages, inflectional markers appear in
patterns that go beyond the simple necessity of satisfying syntactic
requirements/functions contributing to complexity. For example, the different
paradigms drawn for Sanskrit nouns depending on whether it is an a-stem,
i-stem, or u-stem ending word, is an added layer of complexity that is seen as
going beyond functional requirements.

Morphological complexity is an important area of investigation for different
linguistic subfields, and this volume will prove useful to many of them.
Typologists will find the structural resources employed in the various
languages discussed in the chapters interesting. Sociolinguists and those
working with endangered languages would take a keen interest in this
phenomenon due to what morphological complexity or a reduction of a once
complex morphological system has to say about the vitality of the language
(with simplification in structure often linked to a decline in the vitality).
Historical linguists find the principle of compensation, discussed in Chapter
9 in this volume, useful in understanding the diachronic development of a
language. Studying complexity is of great importance for psycholinguists as
well. Their investigations often involve the parsing, processing, assessing,
representation, and production of these complex structures. Many of these
issues are touched upon in the chapters.

Most discussions on complexity and simplification end up being only
impressionistic, with the investigator’s own L1 deciding if another language
is to be treated as having a very complex structure or a moderately complex
one and simplification is often understood in such impressionistic studies as
merely signifying the loss of some erstwhile structures or strategies in the
language. This work must particularly be commended for putting figures to such
impressions and providing tools for furthering the study of morphological
complexity. Reading this volume makes it fairly obvious what Juola (1998)
meant by saying that “to compare, one must first measure, and the process of
measurement is not intuitively obvious” (206).

A major issue with the chapters on ‘Measuring Complexity’, however, is an
over-abundant use, with a kind of presupposed familiarity, of technical terms
and analytical tools that are not usually part of a standard linguistic
training. Thus, the tools used for calculations, and the terminology employed,
make it difficult for the book to be used as a stand-alone resource, as most
readers will need to have some background reading beyond linguistics to
understand and appreciate fully the conceptual and computational tools
employed in the task of ‘Measuring Complexity’. If these chapters on
‘Measuring Complexity’ have been written with a non-linguistic audience in
mind, it has not been made obvious in the introductory chapters.

The authors have been fairly successful in discussing the nature of
morphological complexity in quite a varied set of languages. The focused
treatment of complexity in morphology as delineated from complexity in other
domains of language, helps the scholars in not only explaining how, in spite
of universal parameters of variation, languages come out to be very different
from each other in the domain of morphology unless genetically related, which
therefore rightly merits a treatment different from the treatment of
complexity in syntax or phonology. This focused treatment also enables the
scholars to employ tools that can study complexity at a paradigmatic level.
This is also important because of the role morphological systems play in the
establishment of genetic links between languages. It is indeed intriguing that
while languages unrelated to each other still follow universal
patterns/parameters in their syntactic and phonological organization, the
domain of morphology only allows very macro-level generalizations, like the
use/non-use of inflection, the property of agglutination and its absence, etc.
It is probably because of such lack of micro-level morphological universals,
that morphological complexity is often left to impressionistic
generalizations, as the definition and exact parameters of complexity at the
level of morphology tend to differ quite unsystematically. This volume is
mainly an attempt not to systematize into universal parameters the study of
morphological complexity, but to provide computational tools to measure
complexity. Nichols (1992) had famously quipped when talking of complexity
that “how to measure complexity is itself an issue of some complexity”, and
therefore what the volume achieves with respect to the task of measuring
complexity is no easy task. We are, however, still some way from finding out
the applicational use of these complex measuring techniques.

While no claims for the applicability of the measurement techniques discussed
in the volumehave been made with respect to their usefulness in the general
typology of languages, such an orientation should not have been entirely
ignored. We are certainly aware that complexity can be studied solely in terms
of its calculations and algorithms, but without a larger perspective and
grounding within the linguistic system of variation within defined patterns,
the study remains incomplete even though very important. These issues are
being raised as possible areas for future research and not as in any way
taking away from this very insightful and stimulating book . The volume does
provide useful insights to various sub-disciplines in linguistics.

Complexity is also a matter of linguistic rarity. If linguistic features were
found in all languages, we would probably not be talking of them as being
‘complex’. However, this rarity also raises an important question with respect
to the architecture of language as well as with respect to the human capacity
for language or UG. That children who speak these languages acquire the most
complex of these systems with ease makes Anderson (in this volume) wonder if
morphological complexity is deeply ingrained in the nature of language. This
will be a very fruitful area of investigation in the future, once we measure
and prove that certain structures are more complex or that certain languages
have more complex morphological structures than others. An explanatory
approach to complexity promises to be very challenging in our current state of
knowledge.

If complexity is an important typological parameter then it becomes important
“to map the boundaries of that range according to some metric” (Bane 2008:
69). A fruitful dimension for further research could be to investigate
typologically not only the typological patterns in complex structures, but
also the dominant patterns of simplification when it occurs, and to see if
there are any universal systems underlying simplification of complex
structures. Psycholinguists also take keen interest in morphological
complexity with respect to its effect on short-term memory, especially whether
complex words prove indifferent in the chunking process. Computational and
neuro-linguists are interested in knowing how complexity is automated. Some of
these issues are touched upon in the chapters, while most are left out.

This volume is a useful beginning in our attempt to concretize the conceptual
understanding of complexity. That there aren’t many book-length treatments of
morphological complexity makes this volume not only a path-breaking attempt,
but also a very important contribution to knowledge. Readers from different
orientations in linguistics will find this book useful and rewarding.

REFERENCES

Bane, Max. 2008. ''Quantifying and Measuring Morphological Complexity.'' In
Proceedings of the 26th West Coast Conference on Formal Linguistics, ed. by
Charles B. Chang and Hannah J. Haynie. Somerville, MA: Cascadilla Proceedings
Project, 69-76.

Dressler, Wolfgang U. 1999. ''Ricchezza e complessità morfologica.'' In 
Fonologia e Morfologia dell’Italiano e dei Dialetti d’Italia, ed. by Paola
Benincà and Nicoletta Penello. Rome: Bulzoni, 587-597.

Juola, Patrick. 1998. ''Measuring linguistic complexity: The morphological
tier.'' Journal of Quantitative Linguistics 5.3: 206–13.

Newmayer, Frederick J., and Laurel B. Preston. 2014. Measuring grammatical
complexity. Oxford: OUP.

Nichols, J. 1992. Linguistic diversity in space and time. Chicago, IL:
University of Chicago Press.


ABOUT THE REVIEWER

I work on the morphosyntactic structure of Indian languages from a typological
and areal perspective. I have worked on typological aspects of the
Austroasiatic languages, especially on different Mon-Khmer languages spoken in
Meghalaya in the Northeastern region of India and have a Doctoral thesis on
''The typology of clitics in the Austroasiatic languages of India.'' I am
currently teaching at the English and Foreign Languages University, Hyderabad,
India. My career interests include working on the morphosyntax of
lesser-studied languages of India from a typological perspective.





------------------------------------------------------------------------------

*****************    LINGUIST List Support    *****************
                       Fund Drive 2016
Please support the LL editors and operation with a donation at:
            http://funddrive.linguistlist.org/donate/

This year the LINGUIST List hopes to raise $79,000. This money 
will go to help keep the List running by supporting all of our 
Student Editors for the coming year.

Don't forget to check out Fund Drive 2016 site!

http://funddrive.linguistlist.org/

For all information on donating, including information on how to 
donate by check, money order, PayPal or wire transfer, please visit:
http://funddrive.linguistlist.org/donate/

The LINGUIST List is under the umbrella of Indiana University and
as such can receive donations through Indiana University Foundation. We
also collect donations via eLinguistics Foundation, a registered 501(c)
Non Profit organization with the federal tax number 45-4211155. Either
way, the donations can be offset against your federal and sometimes your
state tax return (U.S. tax payers only). For more information visit the
IRS Web-Site, or contact your financial advisor.

Many companies also offer a gift matching program, such that
they will match any gift you make to a non-profit organization.
Normally this entails your contacting your human resources department
and sending us a form that the Indiana University Foundation fills in
and returns to your employer. This is generally a simple administrative
procedure that doubles the value of your gift to LINGUIST, without
costing you an extra penny. Please take a moment to check if
your company operates such a program.


Thank you very much for your support of LINGUIST!
 


----------------------------------------------------------
LINGUIST List: Vol-27-2749	
----------------------------------------------------------
Visit LL's Multitree project for over 1000 trees dynamically generated
from scholarly hypotheses about language relationships:
          http://multitree.org/