Bauer, Laurie (2001) Morphological Productivity. Cambridge
University Press, xiii+245pp, hardback ISBN 0-521-79238-X,
$59.95, Cambridge Studies in Linguistics 95.
Dorota Smyk, University of Basel
[This book was previously reviewed at
http://linguistlist.org/issues/12/12-2882.html --Eds.]
In "Morphological Productivity", Bauer presents an overview of
the research in the field of morphological productivity. He cites
all the major developments in the field, and provides an in-depth
analysis of the selected nominalization processes. In writing
this book his stated goal was to provide a wide and clear
perspective on morphological productivity that can be used as a
basis for future advances in this field. The book consists of 7
chapters, each ending in a brief summary, complemented by a list
of references and two indexes (language index and subject index)
and lists of figures and tables. Chapter 1 (Introduction) starts
by pinpointing problems resulting from terminological differences
between 'productivity' and 'creativity'. It offers a preliminary
discussion of the potential for the creation of new words.
Through this discussion, Bauer raises a number of questions,
which he attempts to answer in the subsequent chapters.
In Chapter 2, (A histographical conspectus) Bauer presents the
spectrum of different approaches to, and initial problems with,
the notion of productivity, as discussed in literature on the
topic. The problems can be grouped according to the four main
questions: how, where, when and what can be productive. The
chapter discusses:
- degrees of productivity (absolute and cline approaches);
- domains of application of productivity (as it is observable in
certain kinds of word formation);
- the role of lexicalization;
- the difficulty in distinguishing between synchrony and
diachrony;
- the difficulty in distinguishing between competence and
performance.
The six types of definition of productivity in the current
literature collected by Rainer (1987) are also summarized, and,
for the purpose of discussion, the following general working
definition is adopted: "A morphological process is productive if
it can be used to coin new words." (page 27)
Chapter 3 (Fundamental notions) contains a direct elaboration on
the concepts referred to in Chapter 2. Other variables, which
might help in validating and making the working definition of
productivity more precise, are also studied. Here Bauer provides
an exhaustive analysis of a variety of notions, based on the
available literature. He uses many examples to support his
analysis. Altogether, he considers a long list of terms, which
are defined and discussed in detail. The chapter starts with the
notion of existing, new, and potential words, and the criteria
for their classification. Bauer discusses the types and degrees
of lexicalization, introducing the concepts of semantic and
phonological lexicalization. Next the interaction of productivity
with frequency discussed. The distinction between type- and
token- frequency is highlighted, in relation to markedness and
lexicalization. After presenting transparency, attestation, and
default, Bauer attempts to resolve the conflicting
interpretations of 'productivity' and 'creativity'. He proposes
to treat both as hyponyms of innovation with rule-governedness as
the distinguishing criterion. He also discusses the influence of
paradigmatic forces on the coining of new words. The chapter
closes with an exhaustive analysis of the notion of analogy.
Here, morphology is discussed from the perspectives of rule-
governedness and analogy, with no clear distinction between
productive formations and analogical formations emerging.
Therefore, co-existence of the two systems is postulated with
"rule-governed productivity ... [as] a better research strategy"
(page 97) as it allows the discovery of "hidden regularities"
(ibid.) and offers support of analogy in cases where rules fail
to apply. On the basis of the discussion in this chapter,
productivity of a morphological process is redefined as "its
potential for repetitive non-creative morphological coining."
Chapter 4 (Psycholinguistic evidence about productivity) examines
the psycholinguistic evidence about storage of complex words, and
the insight it gives into production and comprehension.
Inflection, derivation and compounding are discussed in turn. The
evidence that seems to support the theory that the most frequent
morphologically complex words may be stored as whole lexical
items, with the most complex words being stored in terms of
morphemes.
Chapter 5 (Scalar productivity) presents qualitative and
quantitative approaches to productivity. First, a number of views
on scalar productivity are summarized. Then, limitations to
productivity are analyzed. He considers the ten general
structural restrictions listed by Plag (1999) as not being as
crucial as detailed constraints of phonological, morphological,
syntactic, semantic, lexical, pragmatic and aesthetic nature as
well as blocking. Base- driven and affix-driven constraints are
distinguished but this distinction is not seen as crucial to the
present discussion. The conclusion resulting from this discussion
is that the constraints can be subdivided into two groups:
(1) absolute constraints, seen as part of competence;
(2) variable constraints, which are considered less reliable,
constituting part of performance.
Finally, the ways of measuring productivity and whether
productivity can be measured at all, are addressed. The analysis
shows that the best available methods, arising from combinations
of dictionary- and corpus-based methods, are still far from
optimal.
After five chapters of academic discussion (though supported by
numerous examples), in Chapter 6 Bauer illustrates some of the
theoretical points, to verify "what productivity is and how it
should be dealt with." (page 163) The first example, the analysis
of the proto Germanic suffix *-dom on the basis of Danish, Dutch,
English and German, illustrates the diachronic difference in
productivity, and supports the claim that profitability is not
directly related to the number of constraints on a particular
process. With -ness nominalizations of colour words, Bauer looks
at the sources of limits of constraints on the base (familiarity
of the base form, its morphology, etymology, length) and
concludes that they all contribute to the acceptability of -ness
colour derivatives, and thus influence profitability.
Nominalizations of English verbs are grouped into three
categories: minor, moderately successful, and successful types.
All these types are analyzed with respect to their profitability
(corpus- and dictionary based measures are compared). Based on
the experiment, Bauer concludes that "we still have no reliable
measure of productivity" (page 199). At the end of the chapter he
examines the question whether the formation types or paradigm
slots are productive on the example of agentive and instrumental
nominalizations, in -er in English. He draws significant
conclusions about productivity.
Chapter 7 (Conclusion) provides a summary of the answers to the
questions asked at the end of Chapter 1. Bauer elaborates on the
provisional definition in Chapter 3 to propose a formal
definition of morphological productivity. The book concludes by
observing how the notions discovered for morphology apply in
other areas of linguistics, such as phonology and syntax.
In 1999, with the publication of Plag's "Morphological
Productivity" and Bolotzky's "Measuring productivity in word
formation" we could observe the need to systematize and evaluate
the findings about the notion of morphological productivity.
Bauer's book is, unquestionably, an important contribution. He
puts into order a mixed collection of observations in a theory-
free framework. As he himself states in the preface, the book
attempts to "provide a stepping- stone in the development of a
new deeper understanding" of productivity. In my opinion it
succeeds. One of the strengths of the book is the clear
presentation of information, with the gradual analysis of the
variables that may foster or hinder productivity. Instead of
offering a straightforward definition of morphological
productivity, Bauer gradually develops his position throughout
the book. He evaluates the merits of all available theories that
have been put forward on this topic, and postulates his
understanding of productivity based on factors that he finds
crucial. At times the treatment seems deliberately noncommittal -
competing theoretical positions are presented. Therefore, a
number of questions remain only partly answered, and some new
ones are triggered. I would like to point some of them here.
In defining productivity as a combination of two independently
functioning mechanisms, availability and profitability, he seems
to make a neat reformulation of Corbin's view of productivity
(McCarthy 1992:32-38). Unlike Corbin, he postulates a distinction
between language system and language norm. In this way,
productivity defined in terms of availability can be presented as
an absolute notion, whereas profitability is a matter of degree,
in a rule- governed system. This, however, still leaves the
constraints limiting profitability very under-specified, and the
number and types of extra-systemic factors unclear. So the
question why among available, competitive formations (such as
'leadfree' and 'unleaded') one formation gains greater
acceptability, requires further research.
Another important issue discussed in the book is the quality of
the data serving as bases for morphological analysis. The
problems arise in the classification of words as attested and
existent, the choice of a dictionary or a corpus as the source
(for additional discussion of methodological problems, see Plag
1999: 96-103). They are also triggered by the subjective
perception of a synchronic picture of language as well as the
adopted framework. Therefore, it seems to me that there is a
discrepancy between what is seen as available on the system
level, and availability as seen by an average speaker-listener.
This, in turns, leads to differences in perception of the
operation of word formation rules. Word re-analysis, a method
employed by language speakers, needs further investigation and
recognition. This is partly recognized by Bauer when he says "it
could be that the speakers work with analogy, but that linguists'
descriptions of the output of this behaviour are in terms of
rules." (page 97) To illustrate it briefly, let us look at two of
the -able words discussed (page 144). 'Enjoyable' and
'acceptable' are, as Bauer points out, examples of borrowing and
result of word formation, respectively. They are both, however,
analyzable and the fact that 'enjoyable' entered the language
before the process had arisen has different implications for a
linguist and for the average speaker. To what extent this
historical knowledge should be employed in morphological analysis
remains to be established.
The last point I would like to make is the perceived cooperation
of productivity and analogy. While discussing analogy, Bauer
comes to a compromise between rule-governed productivity as an
approach to language investigation, and analogy as a useful tool
for the difficult, problematic cases. Yet, as Bauer himself
notes, the border between the two is fuzzy. It leaves the
treatment of cases such as words ending in -ist, -ism etc. (page
83), unresolved. In some instances they share a free base (race,
racist, racism) but in others they do not (baptist, baptism),
with the precise relation with the existing free base not
necessarily easily specified in each case. He claims that "words
without apparent bases can still be derived by rules" and that
"in particular, the pairs in -ism and -ist do not show that such
words must be created by analogy." (ibid.) So, in the case of -
ist and -ism, we have at least the following possible types of
analysis:
(1) Words in Xist/Xism are independently derived from a free base
X if it exists. The others are derived from one another by
deletion/truncation before another affix is added (with the
preference for Xist being derived from Xism). This gives us, in
fact, two different rules.
(2) Regardless of whether the free base is available, Xist is
always derived using truncation/deletion rule on Xism.
Neither proposal seems satisfactory. The problem becomes more
relevant when we consider the number of such coinings. Leaving
these cases to the domain of analogy, however, means excluding
them from the operation of productive morphological rules as
"[b]y definition, the products of analogical formations are not
attributable to the application of any productive word-formation
rules." (Szymanek 1989:103).
Overall the book is highly readable, methodically arranged, with
clear development of the concepts.
BIBLIOGRAPHY
Bolozky, Shmuel (1999) Measuring Productivity in Word Formation:
the case of Israeli Hebrew. Brill.
Carstairs-McCarthy, Andrew (1992) Current Morphology. Routledge.
Plag, Ingo (1999) Morphological Productivity: structural
constraints on English derivation. Mounton de Gruyter.
Szymanek, Bogdan (1989) Introduction to Morphological Analysis.
Panstwowe Wydawnictwo Naukowe.
ABOUT THE REVIEWER
Dorota Smyk is a research assistant at the Univ. of Basel. She
participates in the SNSF-funded research project "Word Formation
as a Structuring Device of the English and Italian Lexicons: A
Large-Scale Exploration" as a Word Manager lexicographer. At the
same time she is also pursuing a Ph.D. degree in General
Linguistics. Her research interest is focussed on morphological
productivity and the role of lexicons in language-teaching and
learning.
Links:
http://www.unibas.ch/LIlab/staff/dorota/index.html
http://www.unibas.ch/LIlab/projects/wordmanager/IT-EN-Project.html
|
