Claridge, Claudia, (2000) Multi-word Verbs in Early
Modern English: A Corpus-based Study, Rodopi,
Matthew Walenski, Department of Neuroscience and
Linguistics, Georgetown University.
Claridge offers a description of multi-word verbs in English, based on the
Lampeter corpus of 17th - 18th century written English, specifically from
the period 1640 - 1740. Multi-word verbs are defined as analytic
constructions that "nevertheless represent a semantic unity that is
characteristic of a single word or lexical unit (p.26)." Chapter 2
provides a detailed description of the Lampeter corpus, and a discussion
of the problems inherent in corpus-based research. The focus of chapters
3 and 4 is to provide a system of classification for multi- word verbs and
define membership within each class. She divides multi-word verbs into
five categories: phrasal verbs (e.g., find out), prepositional verbs
(e.g., depend on), phrasal-prepositional verbs (e.g., break in on),
verb-adjective combinations (e.g., put straight), and verbo-nominal
combinations (e.g., take a walk), further divided into three types.
Chapter 5 examines the history of these types of multi-word verbs from Old
English through to Modern English. Chapters 6 and 7 describe the
particular multi- word verbs found in the Lampeter corpus, and attempt to
describe the patterns found among them. Chapter 6 treats the data
synchronically, while chapter 7 looks at diachronic developments during
the 100 year period covered by the corpus. Chapter 8 is a review of
contemporary grammatical accounts to ascertain whether grammarians
(prescriptive and no-so-prescriptive) were aware that multi-word verbs
existed in the language, and if so, what their attitude was towards these
forms. Chapter 9 then attempts to describe factors that influence
language choice. Given an option between a multi-word verb and a simplex
verb (e.g., 'take a walk to the store' vs. 'walk to the store'), which
would a speaker choose, and why? At the end of the book is an appendix
that lists all the multi-word verbs Claridge finds in the Lampeter corpus,
according to her system of classification.
First let me emphasize, as Claridge herself states, that the work does not
offer a theoretical account or analysis of multi-word verbs, it is
intended to be purely data oriented. However, adding a theoretical
perspective would perhaps have addressed the book's most serious problem:
it comes across as subjective, unfocussed, and inexplicit. As a result,
she failed to convince me that the data she presents reflects anything
real about this period of English.
The entire book is very well sourced, and the amount of literature
reviewed seems quite comprehensive, though with a couple of notable
exceptions. Two theoretical accounts of these types of predicate in the
literature are Ackerman and Webelhuth (1998) and Di Sciullo and Williams
(1987), though this is certainly not intended to be a comprehensive list.
Ackerman and Webelhuth (1998) provide an account which is explicitly
intended for predicates that may not be simplex forms, and as such is
perfectly suited for the data Claridge presents. Di Sciullo and Williams
(1987) provide an explicit definition of these types of (multi-word)
objects, though their terminology differs slightly. However, no reference
is made to these works or to any of their references (despite many
references more recent than 1998). While this is an unfortunate
oversight, it would be unfair not to mention that there is little overlap
in the references in either direction, so perhaps this is less a fault of
Claridge and more an indication of a general rift between different
schools of linguistics.
Chapter 2 is a pleasure to read, and is a well organized and well
presented summary of the Lampeter corpus and the particular advantages it
offers. Additionally, Claridge includes a lot of very interesting
historical and social commentary on the way in which written information
was used at the time of the writings in the corpus. Chapter 8 recaptures
this same interesting spirit, and is a wonderful slice of grammatical
history - one rarely has a chance to read such a concise, organized
presentation of very interesting material. Chapter 5 was also
interesting, but perhaps not convincing, for reasons to do with her
classification system (see comments below).
However, the rest of the book suffers greatly. First, there is no
explicit definition of the data she is treating: "The notion of 'word'
will throughout the study be used in an everyday, quasi pretheoretical way
as a more theoretical definition is not necessary for the problem at hand
(p26)." Given that this is a book about (multi-) words, one would think
that a precise definition of 'word' would be exactly what was necessary.
Here certainly, the theoretical approaches mentioned above (notably Di
Sciullo and Williams) could have been used to great advantage.
The rest of chapter 3, which attempts to define a system of classification
of these multi-word verbs, appears very subjective. At the end of the
chapter, she states: "we all know in some way what e.g., a phrasal verb
is, but a full and theoretically adequate proof of this intuitive
knowledge seems impossible. If in doubt, I will therefore trust my
intuitions more than I will trust any kind of test (p41)." While Chapter
4 does attempt to provide objective tests to classify predicates, they are
undermined at every turn by her failure to rely on their objective
An additional problem is that the literature review, while fairly
comprehensive (barring omissions mentioned above) is organized in such a
way that it is impossible to evaluate why she rejects some work and not
others, or to know how the system she proposes is superior to these prior
works, if indeed it is so. In fact, I was left with the impression that
she had tried to make a fresh start by more or less randomly sampling from
prior work in this area, taking some ideas and not others for reasons that
were not explicitly or objectively clear. Unfortunately I was not
convinced that this was necessary, nor was I convinced that the system she
proposes is better than those that came before.
To give an example of one of her dismissals: "At the beginning of this
section, I said that [verb-adjective combinations do] not pattern nicely.
But Bolinger ... and Konig ... advocate a possible subdivision into three
groups... While this classification works with many, perhaps even the
majority of verb-adjective combinations, there are cases which are not
covered by it... (p69)."
While the concept of the data that she presents and advocates does have
its appeal, her failure to convince the reader that she has any objective
tests for the phenomena she is studying, and her failure to convince the
reader that she has objectively applied even the tests she has, raise
serious doubts about the validity of her classification (despite its nice
organization in the appendices). Unfortunately, as this serves as the
basis for the remainder of the study (the corpus investigation), the
results of that investigation are cast into doubt as well.
Chapters 6 and 7 constitute her own research with the Lampeter corpus.
The information she presents is somewhat interesting, but any sort of
explicit motivation for the measurements she makes would be helpful, as
there appears to be no rhyme or reason to the data. Second, her use of
statistics is chapter 7 is somewhat odd. She presents the results of
several (not clearly motivated) statistical comparisons (chapter 8 does
finally provide motivation for one of her comparisons, to be fair),
decides "statistical significance is not everything" (p.177) and proceeds
to interpret the data as if the statistical tests had never been done in
the first place. This puts the reader in a tough spot: Does one take the
statistics at face value, and ignore the conclusions she draws? Or does
one ignore the statistics, as she does, and accept her conclusions? Why
were these tests performed, if they have nothing to offer? Moreover, how
is the data affected by the subjectivity of her classifications?
Here again is where some theory would be useful, if only to provide some
explicit reason (other than intuition) to expect a pattern. However,
Claridge states: "This sequence is not surprising, in fact, it is exactly
what I would have expected intuitively ... an intuition which,
unfortunately, cannot at the moment be verified by empirical
studies... (p175)" However, she does not succeed in conveying these
intuitions to her audience, leaving us very little to go on.
The goal of chapter 9 is to describe what factors would cause an author to
use a multi-word verb over a simplex equivalent. In other words, to make
her case she needs to get inside the heads of authors who have been dead
for three hundred years. This is very ambitious, and while she should
perhaps be commended for the attempt, it should not come as a surprise
that she makes little progress. While the introduction to the chapter
warns us of its speculative nature, her speculation in many places goes
beyond what could fairly be considered speculation. To give one example,
when speaking of a metaphoric usage of the verb 'eat,' in the sense of "to
destroy as a parasite or corrosive" (p235n10) she says, "For people living
in early modern times, witnessing such processes would have been a rather
common experience, i.e. the picture was more accessible to them than it is
to us (ibid.)" Certainly this is speculative, but inasmuch as it is used
to support an argument (even a speculative one), how could such a claim
possibly be substantiated?
To summarize, I would recommend this book in two instances. First, to
someone interested in the Lampeter corpus, Chapter 2 would be a very
helpful and useful introduction. Second, to someone deeply committed to
doing research into multi-word verbs, her findings and methods may be
relevant, but should be taken with a grain of salt.
Ackerman, F., and Webelhuth, G. (1998) A Theory of Predicates. CSLI,
Di Sciullo, A.M., and Williams, E. (1987) On the Definition of Word.
MIT Press, Cambridge.
My research interests include sentence processing (psycho- and
neuro-linguistics), syntax, phonetics, historical linguistics, and writing
systems. I am currently a post-doc at Georgetown, working on memory and