LINGUIST List 11.2729

Fri Dec 15 2000

Review: de Monnink: On the move

Editor for this issue: Andrew Carnie <carnielinguistlist.org>


What follows is another discussion note contributed to our Book Discussion Forum. We expect these 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 discussion." (This means that the publisher has sent us a review copy.) Then contact Andrew Carnie at carnielinguistlist.org

Directory

  1. j.mukherjee, Review of de M�nnink (2000)

Message 1: Review of de M�nnink (2000)

Date: Fri, 15 Dec 2000 09:18:09 +0100
From: j.mukherjee <j.mukherjeeuni-bonn.de>
Subject: Review of de M�nnink (2000)

Inge de M�nnink (2000): On the move: the mobility of constituents in the
English noun phrase - a multi-method approach (Language and Computers:
Studies in Practical Linguistics No 31). Amsterdam: Rodopi.

Reviewed by Joybrato Mukherjee, University of Bonn


The present work is based on the author's PhD thesis defended at the
University of Nijmegen. It clearly stands in the tradition of Nijmegen
linguists such as Jan Aarts who have certainly been in the vanguard of the
development of English corpus linguistics over the last decades. However,
Inge de M�nnink does not only explore the positional mobility inherent in
the English noun phrase (NP) from a genuinely corpus linguistic perspective,
but she also draws on intuition-based elicitation data. The combination of
those two approaches is what de M�nnink refers to as "multi-method
approach". Thus, it is not only the object of inquiry as such, but also the
methodology which is central to the present review and warrants a detailled
description.


Synopsis

The organisation of the book is as follows. Chapter 1 provides an overview
of some previous approaches to the English NP. In chapter 2, de M�nnink
gives reasons for the multi-method approach she favours and describes both
the corpus analysis and the elicitation experiment. The results of the
corpus analysis are then presented and discussed in chapter 3. The corpus
findings provide the framework for the elicitation experiment the results of
which are outlined in chapter 4. Reviewing some previous attempts to explain
the mobility of constituents in the English NP, de M�nnink sets out to
explain her findings with the help of both formal and functional principles
in chapter 5. These principles may lead NP constituents to be moved to the
left or to the right. Chapter 6 brings together the linguistic evidence and
the methodological considerations and nicely rounds off the book with a
critical evaluation and some prospects for future research. The rich
appendix gives all the necessary information on the corpus design,
abbreviations, results of the quantitative corpus analysis and the
elicitation experiment.

In the opening chapter, de M�nnink discusses some previous approaches to the
English NP. She starts off by reviewing the classic reference grammars by
Poutsma (1904-26), Kruisinga (1909-32) and Jespersen (1909-49). With regard
to more recent research, special emphasis is put on Quirk et al. (1985),
Aarts/Aarts (1988) as well as on structural, functional and generative
grammar. From the multitude of these approaches, the author filters out the
so-called "prototypical NP structure" (p. 19):

-(limiter): adverb phrase
-(determiner): determiner phrase
-(premodifier)*: adjective phrase or adverb phrase or noun phrase
-head: noun or pronoun or proform
-(postmodifier)*: prepositional phrase or clause

Brackets indicate optionality, asterisks mark possible multiple
realisations. It should be noted that this scheme clearly distinguishes
between functions of constituents (e.g. limiter) and their formal categories
(e.g. adverb phrase), thus following the original Nijmegen approach to
corpus annotation which, for example, has also been applied to the
International Corpus of English (ICE). The prototypical NP structure is
subject to alteration in so-called "variant NPs". In this, any constituent
may be moved to the left ("fronted"), to the right ("deferred") or even be
placed outside its mother constituent ("floating", either fronted or
deferred). On this basis, nine types of variant NPs are classified, e.g.
"NPs with a deferred determiner" as in "Make it quite clear to us all" (p.
27): here "all" is the deferred determiner in the NP "us all". The focus of
the present work is clearly on the frequency, distribution and structural
variety of such variant NPs.

In the second chapter, some highly inspiring and thought-provoking
methodological considerations are presented. De M�nnink makes it clear at
the outset that in spite of its objectivity and verifiability a merely
corpus-based approach to the description of the English NP has two
disadvantages: (1) there is no corpus which covers all possible NP
structures to a sufficiently large extent; (2) it remains unclear whether
constructions not occurring in a corpus are in fact
unacceptable/ungrammatical, i.e. whether it is due to the corpus size or due
to the language as a whole that a certain NP structure is not attested. In
this context, she largely capitalises and elaborates on other corpus
linguists' attempts to reconcile corpus-based methodology with
intuition-based descriptions of language (e.g. Aarts 1991). Accordingly, de
M�nnink suggests to complement her corpus-based study with intuitive data
which could serve to enlarge the range of linguistic structures, e.g. the
versatility of variant NPs. The author does, however, not confine herself to
the object of inquiry at hand, but develops a very innovative and general
"data cycle for descriptive linguistics" which could be applied to the
exploration of other linguistic phenomena as well (p. 34):

1. literature => hypothesis/-es
2. corpus data
3. literature => hypothesis/-es
4. intuitive data
(back to 1. => cycle completed)

This cyclical procedure takes into account, for example, that even
corpus-based studies are wont to be implicitly based on some intuition-based
hypotheses of the linguist derived from, say, previous studies (i.e.
literature) or some own experience. De M�nnink points out that a descriptive
linguistic study might start anywhere on the data cycle "as long as the
whole round is completed at least once" (p. 34). The catch, though, is that
such a data cycle should not end with a simple introspection of the
linguist, which would infringe the empirical grounding of the previous
analytical steps. Rather, what de M�nnink suggests is a well-defined
elicitation of intuitive data which allows for a more exhaustive grammar
construction than would be possible by relying on corpus data alone. In the
remainder of this chapter, she sketches the design of the two corpora she
uses (comprising 170,000 words and 220,000 words respectively) and the
design of the elicitation experiment. The elicitation experiment is based to
a large extent on the design used at the Survey of English Usage (SEU) and
draws on both performance and judgment tests. These tests were presented to
45 native speakers in the first round and 75 persons in the second round.
For example, they were asked to perform a preference rating of the sentences
"No more was heard of him" and "No more of him was heard" along a
seven-point scale ranging from "perfectly acceptable" to "totally
unacceptable" (p. 48). It should be noted that de M�nnink explicitly (and
correctly, I believe) subscribes to the view that such elicitation data
reflect a speaker's language behaviour and not - as some linguists suggest -
his or her internal grammar or competence.

The third chapter presents in great detail the outcome of the corpus
analysis and gives many authentic corpus examples. It is beyond the scope of
the present review to replicate all results. In general, it is most
interesting to see that variant NPs which do not follow the prototypical NP
structure are not at all a marginal phenomenon. De M�nnink finds 1,287
examples in the 170,000-word corpus, thus representing 5.3% of all complex
NPs, i.e. NPs which do not consist of a head only (p. 58). Of all types of
variant NPs, quite a few, however, occur only sporadically in the corpus
material. For example, the NP type "deferred modifier + discontinuous
determiner" occurs only twice in the 170,000-word corpus (more specifically,
in the genre of scripted speech). Such low frequencies pose of course
serious problems for the reliability of statistical analyses such as the
chi-square test. Furthermore, NPs with a fronted premodifier exclusively
function as subject complements, direct objects and prepositional
complements in this corpus. Also other types of variant NPs are not attested
in one clause function or the other. These findings refer to the problems of
corpus size and the lack of coverage of language use in its entirety (see
above). Therefore, de M�nnink draws the conclusion that three types of
variant NPs should be in the focus of elicitation experiments with native
speakers: the fronted premodifier, the discontinuous adjective phrase and
the floating deferred modifier. Following the data cycle, some corpus-based
hypotheses as to the acceptability of those types of variant NPs are put
forward at the end of the chapter, e.g. "An NP with a fronted premodifier
cannot function as indirect object" (p. 83). These hypotheses are to be
tested in the elicitation experiment. Finally, I would like to point out
that this chapter also offers an interesting discussion of some minor issues
of word categorisation. For example, "such" and "other" are subsumed into
the functional group of premodifiers and the word class of adjectives (p.
76). Albeit at odds with some previously suggested categorisations of those
items, de M�nnink's line of argumentation seems entirely plausible to me.

Chapter 4 focuses on the elicitation experiment. After describing in detail
the experimental procedure (including problems encountered while conducting
them), the corpus-based hypotheses are evaluated in the light of the
elicitation data. For example, NPs with a fronted premodifier turn out to be
acceptable in indirect object position according to most native speaker
informants: "You should not tell so jealous a boy that you met his
girlfriend in the pub" is generally not considered as ungrammatical (p. 94).
In a similar vein, all other corpus-based hypotheses are put to the test in
the elicitation experiment. The two sets of data are then combined in order
to formulate new hypotheses. As far as variant NPs with a fronted
premodifier are concerned, for example, the author arrives at the conclusion
that "these NPs are also acceptable in indirect object, object complement
and adverbial position, although such occurrences were not encountered in
the corpus" (p. 108). By incorporating the elicitation data, the descriptive
framework provided by the prototypical NP structure is extended step by step
so that de M�nnink is able to offer a fairly complete description of the
three types of variant NPs under scrutiny. At the end of this chapter, the
author outlines some prospects for a continuation of the data cycle. From
her point of view, it makes sense to search larger corpora such as the
100,000,000-word British National Corpus (BNC) for specific variant NP
types. This would help to corroborate the results from much smaller corpora
and from the elicitation of intuitive data. In retrospect, the sample
searches she made in the BNC confirms the earlier observation that a larger
corpus would most presumably shed new light on the range of possible variant
NPs. To pick out but one example, the BNC data on variant NPs with a
premodifier make the author "include a prototypical premodifier position
which can be realized by a classifying NP or AJP", as for example in: "I had
so romantic a long evening yesterday" (p. 114). Quite obviously, the data
cycle is a genuinely heuristic method of linguistic analysis which, in fact,
has no clear-cut point of completion.

Chapter 5 emphasises the wider issue of mobility of constituents in the
English NP. Here the author addresses the question as to why variant NPs
with mobile (i.e. fronted or deferred) and discontinuous (i.e. floating)
constituents occur in the first place. Her remarks are suggestive rather
than exhaustive. This is due to the fact that a systematic analysis of the
principles underlying the mobility of all constituents in all variant NPs
does not lie at the heart of her study. It is rather the range of NP types
that she wants to describe as completely as possible. Nonetheless, de
M�nnink succeeds in providing a very perceptive overview of some important
theoretical models which seek to account for the mobility of constituents in
general (e.g. movement rules in generative grammar, Hawkins' (1994) "Early
Immediate Constituents Principle" and Kayne's (1994) stranding analysis of
extraposition out of NP). She sets out to evaluate the explanatory power of
these approaches in the light of the corpus and elicitation data on variant
NPs presented in previous chapters. Particular attention is paid to the
three types of variant NPs which have already been in the focus of the
corpus analysis and the elicitation experiment. Generally speaking, it turns
out to be necessary to include both syntactic and pragmatic principles into
a comprehensive theory of mobility.

The last chapter brings together the two main aspects of this book: for one,
the linguistic analysis of the English NP structure is summarised, and on
the other hand, the multi-method approach is critically evaluated. These
considerations include linguistic and methodological issues that future
research will need to address.


Critical evaluation

At the beginning of the summary, de M�nnink recaps the two main goals of her
study: "The first goal was to investigate the mobility of the constituents
in the English noun phrase and gain insight into the nature and frequency of
variant NPs. The second goal, which follows from the first, was to develop a
multi-method approach to descriptive studies that combines corpus data and
experimental data" (p. 147). In my opinion, both of these goals are
unreservedly achieved. This book provides a careful and considered analysis
of variant NPs and, perhaps more importantly, breaks new ground in tearing
down the invisible wall between corpus linguists and "armchair linguists"
(Fillmore 1992). That the combination of corpus data and elicitation data is
in fact useful for the purpose at hand is vindicated by the quantity and
quality of interesting data the author obtains and by the conclusions she is
able to draw from their analysis. This is not to say, though, that I agree
with the author on all details of analysis, methodology and terminology. In
particular, I am not so happy with the way de M�nnink uses the term
"descriptive linguistics". De M�nnink seems to suggest that "descriptive
linguistics" only refers to the exploration of the complete range of
structural possibilites in language, which, of course, will never be covered
by a corpus however large it may be. I personally would also consider a
merely corpus-based description of attested and probable linguistic
phenomena as truly descriptive. (I would like to invite the author to
comment on this point, by the way.) Apart from such minor disagreements, I
have no problems whatsoever with the general procedure and the conclusions
drawn from the corpus analysis and the elicitation experiment. Rather (and
to say the least), I am thrilled at the wealth of interesting data, the
plausibility of arguments and the innovative methodology. I would like to
add that the book is a pleasure to read also because of its style. It is
well written and makes use of many discourse organisers, thus tying together
all the different parts. Throughout the study, the reader is never allowed
to lose track of the main line of argumentation. The book is also well
produced and the proof reading turns out to have been almost perfect so that
only very few errors remain, e.g. the wrong percentage of variant NPs in
Table 3-2 on page 58 (which should be 2.36% instead of 2.63%). Also some
abbreviations are not included in the appendix, e.g. PC for prepositional
complement. This book no doubt makes for stimulating reading, not only for
corpus linguists but for all grammarians interested in the structural
variation at the level of noun phrases. Moreover, both the open-minded
linguist from the world of corpus linguistics and the defender of
intuition-based methods will certainly profit from reading this book in
order to get the best of both worlds. No-one interested in an unbiased
approach to linguistic description in general and/or the English noun phrase
in particular should miss this excellent book.


References

Aarts, F. and J. Aarts (1988): English Syntactic tructures: Functions and
Categories in Sentence Analysis. 2nd edition. New York: Prentice Hall.

Aarts, J. (1991): "Intuition-based and observation-based grammars", English
Corpus Linguistics: Studies in Honour of Jan Svartvik, ed. K. Aijmer and B.
Altenberg. London: Longman. 44-62.

Fillmore, C. (1992): "'Corpus linguistics' or 'computer-aided armchair
linguistics'", Directions in Corpus Linguistics: Proceedings of Nobel
Symposium 82: Stockholm, 4-8 August 1991, ed. J. Svartvik. Berlin: Mouton de
Gruyter. 35-60.

Hawkins, J. (1994): A Performance Theory of Order and Constituency.
Cambridge: Cambridge University Press.

Jespersen, O. (1909-49): A Modern English Grammar on Historical Principles.
Copenhagen: Munksgaard.

Kayne, R. (1994): The Antisymmetry of Syntax. Cambridge, MA: MIT Press.

Kruisinga, E. (1909-32): A Handbook of Present-day English. Groningen.

Poutsma, H. (1904-26): A Grammar of Late Modern English. Groningen:
Noordhoff.

Quirk, R. et al. (1985). A Comprehensive Grammar of the English Language.
London: Longman.



Joybrato Mukherjee is an Assistant Professor of Modern English Linguistics
at the English Department of the University of Bonn. His research interests
include corpus linguistics, stylistics, textlinguistics, syntax, intontaion
and EFL teaching. In his forthcoming book "Form and Function of
Parasyntactic Presentation Structures", interactions between prosody and
syntax at tone unit boundaries are described on the basis of quantitative
and functional corpus analyses.

Email: j.mukherjeeuni-bonn.de
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue