Publishing Partner: Cambridge University Press CUP Extra Publisher Login
amazon logo
More Info

New from Oxford University Press!


Style, Mediation, and Change

Edited by Janus Mortensen, Nikolas Coupland, and Jacob Thogersen

Style, Mediation, and Change "Offers a coherent view of style as a unifying concept for the sociolinguistics of talking media."

New from Cambridge University Press!


Intonation and Prosodic Structure

By Caroline Féry

Intonation and Prosodic Structure "provides a state-of-the-art survey of intonation and prosodic structure."

Review of  The Morphology of English Dialects

Reviewer: K Aaron Smith
Book Title: The Morphology of English Dialects
Book Author: Lieselotte Anderwald
Publisher: Cambridge University Press
Linguistic Field(s): Historical Linguistics
Issue Number: 21.2486

Discuss this Review
Help on Posting
Author: Lieselotte Anderwald
Title: The Morphology of English Dialects
Subtitle: Verb-Formation in Non-standard English
Publisher: Cambridge University Press
Year: 2009

K. Aaron Smith, Department of English, Illinois State University


Anderwald's (henceforth A) monograph is a functionally-oriented treatment of
English verbal morphology, focusing especially on a handful of verbs whose
dialect history differs from that of Standard English.

Chap. 1 Introduction. The introduction sets up some of the important
terminology and argumentation for the remainder of the book, including the
justification for using the terms strong and weak (as opposed to the more
expected regular~irregular) and a classification of verb patterns based on
formal identities among the three so-called principal parts, i.e. the
infinitive, the simple past and the past participle. (In Chapter 3, these
patterns are also expanded to include their various subtypes and to show the
overlapping features among them.) These patterns are further ranked in terms of
their functionality. These identity patterns are: 1. infinitive, simple past
and past participle are all different in form (e.g. 'write', 'wrote', 'written'
[this is also said to be the prototypical strong verb pattern]); 2. infinitive
has a different form from the simple past and past participle (e.g. 'bring',
'brought', 'brought' or 'live', 'lived', 'lived' [note that all weak verbs
belong to this type and it is said to be the prototypical weak verb pattern]);
3. infinitive and past participle have an identical form distinct from the
simple past (e.g. 'come', 'came', 'come'); 4. infinitive and simple past share
a form distinct from the past participle (e.g. 'beat', 'beat', 'beaten'); and 5.
infinitive, simple past and past participle all share the same form (e.g. 'cut',
'cut', 'cut').

The functionality of each pattern has to do with its expressive potential and
avoidance of morphosyntactic ambiguity. Thus, pattern 1 is maximally
functional, since it shows a different form for each of the verb paradigm parts.
Pattern 2, however, is also quite functional in this line of thinking because
although the simple past and past participle are identical, that identity is
disambiguated through syntactic patterning (the past participle, when it appears
in a finite verb phrase does so with auxiliary support). It follows then that
those patterns in which the infinitive and simple past share an identity would
be the least functional. Thus pattern 3 is still ''functional'', whereas patterns
4 and 5 are not. As A shows, however, functionality in these terms does not
fully account for the distribution of strong verbs across these patterns:
Pattern 1 contains 35% of English strong verbs, Pattern 2, 48%, Pattern 3, 1%,
Pattern 4, .05% and Pattern 5, 14% (numbers based on a slightly modified list of
the ''250 or so'' strong verbs listed by Quirk et al. 1985). While functionality
in terms of identity of form correlates well with the high number of verbs in
patterns 1 and 2, the low occurrence of verbs in pattern 3 runs counter to such
functionality-based expectations and the 14% of verbs in pattern 5 is somewhat
surprising since identity among all three forms is maximally non-functional.
Thus, functionality, in so far as it has to do with identity in form, is not the
only explanation in the formal patterning of verbs, although it is one parameter
of explanation.

A's study is socio-historical in scope and in Chapter 1 she lays out the various
diachronic and dialectal corpora from which she is able to glean information,
including: The Oxford English Dictionary, the Helsinki Corpus, ARCHER,
Linguistic Atlas of Late Mediaeval English (presumably vol. 4, the county
dictionary), 'English Dialect Dictionary' and 'English Dialect Grammar', the
'Survey of English Dialects', the Freiburg English Dialect corpus
(, last
accessed 6/6/10-note that the link given in A in footnote 21 of the introduction
did not work when I tried to access it), and the Corpus of London Teenage Speech.

Chap. 2 Past tense theories. Chapter 2 offers a critical review of theories on
past tense morphology, from Chomsky and Halle's 1968 The Sound Patterns of
English, Lexical Phonology and Morphology, Optimality Theory and Stochastic
Optimality Theory, psycholinguistic theories, connectionist approaches, Bybee's
Network Model, and finally Natural Morphology. In her work, A provides good
summaries of these relevant theories. Ultimately the explanations for much of
the development and most of the maintenance of non-Standard verbal morphology is
based on productive patterning as captured in Bybee's functionalist Network
Model (Bybee 1985, 1988, 1995) and Natural Morphology, both in terms of
universal naturalness (Mayerthaler 1988) and language-specific naturalness
(Wurzel 1984).

Chap. 3 Naturalness and the English past tense system. This chapter expands the
notion of morphological naturalness, setting up and then testing some
predictions of diachronic development according to naturalness claims.
Specifically it is predicted that: 1.) a strong verb may become a weak verb; 2.)
a strong verb may take a weak suffix, without identity of the infinitive stem
vowel throughout the paradigm (e.g. 'wroted'); 3.) a strong verb may take over
a dominant strong verb marking; and 4.) there may be an abstract change that
brings about dominant patterning in which the infinitive has a different
identity from the simple past and past participle in a way other than 1, 2, and
3. Naturalness alone is not sufficient for predicting change, however, because
option #2, although historically attested, is quite rare. Chapter 3 concludes
with demonstrations of the historical trends predicted by 1, 2 and 4.

The next three chapters contain case studies on the development of non-Standard
verbal morphology, in which individual histories of some verbs are given. As
the names of the chapters indicate, the studies are grouped by changes toward a
certain pattern: Chap. 4 'Sellt' and 'knowed': non-standard weak verbs; Chap. 5
Drunk, seen, done, and eat: two-part paradigms instead of three-part paradigms;
and Chap. 6 'Come' and 'run': non-standard strong verbs with a one-part
paradigm. In many cases we find that non-Standard morphology is retentive of
much older patterns, although in other cases dialect forms are newer in terms of
historical development. The changes themselves, however, are sometimes
explained by naturalness/functionality, and at other times as conforming to a
sub-regularity, dubbed in A as Bybee verbs (verbs with a mid, central vowel in
the past tense, cf. 'spin~spun').

Chap. 7 Conclusion: Supralocalization and morphological theories. In this
concluding chapter, A sums up that for the most part there are good historical
reasons for the development of non-standard forms, and that their resilience to
change is due to their functionality. This is not to say that functionality per
se is the motivation for their form; instead their form is accidental and
idiosyncratic but may ''become reinterpreted in functional terms'' (187). The
concluding chapter also gives hypotheses concerning the direction of change both
in terms of sociolinguistic and functional criteria; however as A points out,
these are working hypotheses and await further evaluation, an endeavor impeded
in part due to the lack of databases upon which such hypotheses might be tested..


Overall A has brought forth a very good account of morphological patterning for
some varieties of English. Her argument is strongly set out in the first two
chapters, her presentation of verb patterns and their functionality being
especially cogent. Additionally, her review of various theories of verbal
morphology offers a useful summary. In the main, her theoretical argumentation
and findings are convincing. Just a few questions remain, and these do not in
any significant way mitigate her overall thesis.

There appears to be a slight terminological problem involving the consistent
labeling of verb types. In the first chapter, A says that she will be using the
terms strong and weak, even though they can be admittedly troublesome. On page
3 she says that any irregular verb will be called a strong verb. Then she adds
on page 6 that verbs such as 'spill', in which the dental suffix has devoiced,
e.g. 'spillt', have become irregular. Thus, I would take such verbs to be
strong in her terminology. But around p. 64 where she discusses the past forms
'sellt' and 'tellt', she calls them weak. This terminological discrepancy does
not really impact her argument, since the functional explanation is based on the
verb patterns outlined in Chapter 1 (see above), and thus it doesn't really
matter whether one says that these verbs have become weak or whether they have
conformed to a dominant pattern whereby the present has a different form from
the identical simple past and past participle form (although her notion that
'sellt' was the model for 'tellt' (p. 74) is unconvincing given that 'tellt' is,
and presumably has been, more frequent than 'sellt').

As mentioned above, one of the strengths of this work is its comparison of
several databases in order to tease out diachronic patterns. However, in some
cases the use of COLT (Corpus of London Teenage Speech) must be considered with
caution. And although A uses COLT very responsibly and never overstates its
role in the overall picture, there is one aspect of its use that deserves
mention. On page 96 for instance COLT is consulted to test whether certain
non-standard weak forms are prevalent in the present-day verb system. Given
that only one ''non-standard'' weak verb appears in that data, A concludes that
such weak forms in general are not part of the present-day system. This may be
true, but of course more data would be necessary for such a conclusion. Such a
use of COLT assumes an apparent time construct (e.g. Bailey, Wikle, Tillery and
Sand 1992) but as is well known, age-graded data can give false diachronic
information. It may well be, and is often enough the case, that younger
speakers will show greater use of standard forms because they are ''close'' to the
standard due to schooling. As such speakers get older, they sometimes conform
to their speech communities in ways that make such non-standard features more
stable than they might at first appear. Again A doesn't overstate the
importance of the COLT data, but given the current popularity of London-based
youth speech, A's findings clearly suggest much important research left to be
done on the linguistic practices found in COLT in regards to non-standard usage
and its potential spread.

A’s focus is on varieties of British English and it is clear, particularly to a
reader familiar with American English, that A’s hypotheses would be further
elucidated through the inclusion of data other than those from British English.
For instance, the non-standard past form of 'drink' as 'drunk' is motivated in
A as being attracted to the pattern of Bybee verbs, defined above. However
perhaps because of the more or less exclusive focus on British English, A
doesn’t mention that for some speakers the past tense form 'drank' has extended
to the past participle use; in fact it is actually rather frequent in American
English. Not only is it ubiquitous in my speech community of origin, but it
pops up several times through a Google search, with several of the sites
discussing the correctness of 'drank' as a past participle, discussions that
occur precisely because 'drank' is a past participle for many American English
speakers. In fact it is frequently the case that students who are
native-speakers of American English become incredulous to learn that 'drunk' is
the prescriptively correct past participle form! Such data is not really
problematic in the bigger theoretical picture that A builds; after all 'drink',
'drank', 'drank' is functional in A’s terms. In a similar way A slightly
overstates the functional split between 'done' and 'did', in so far as she
claims that the former, when used non-standardly, is only used as a main verb
and the latter only as an auxiliary (p. 126). While I believe A’s point to be
that uses such as ''Done he leave yesterday?” are not predicted, the way the
distribution is stated, it would also not predict 'has did'. However, 'has did'
is well-attested in some varieties of American English, and probably varieties
of British English too.

As a final note, it's rather confusing when A claims that Bybee's Network Model
doesn't predict the direction of language change or when she says that it isn't
diachronic. (pp. 40, 47-48). Neither of these statements is exactly true. The
Network model has been invoked a number of times to explain change, even by A
herself. For instance, Bybee 1985: 130 says of the Network pattern of ''Bybee
verbs'' (note that Bybee herself doesn't call the set as such) that it ''attracts
new members'' (and see A p. 99 where she also says of the pattern that it ''can
attract new members''). Attraction of new members is predictive of a diachronic
change and so it's mysterious to me why the more diachronic aspects of Bybee's
Network model are glossed over when it is first introduced in Chapter 2.

My overall evaluation is that A's arguments are well made, and often quite
convincing. What gives her approach such theoretical strength is the comparison
of data from several sources and types of sources, diachronic and dialectal. To
my mind, this is one of the most fruitful ways to engage diachrony in linguistic
explanation since synchronic distribution/variation is a result of diachronic
events. In this way, A's study adds to the still-too-scarce approach whereby
diachrony is included in theoretical accounts that are not merely historical in


Bailey, Guy, Tom Wilke, Jan Tillery and Lori Sand. 1992. ''The apparent time
construct''. Language Variation and Change 3, 241-264.

Bybee, Joan L. 1985. Morphology: A Study in the Relationship between Meaning and
Form. Amsterdam: Benjamins.

Bybee, Joan L. 1988. ''Morphology as lexical organization''. In Michael Hammond and
Michael Noonan, eds. Theoretical Morphology: Approaches in Modern Linguistics.
San Diego: Academic Press, 119-141.

Bybee, Joan L. 1995. ''Regular morphology and the lexicon''. Language and Cognition
Processes 10, 425-455.

Mayerthaler, Willi. 1988. Morphological Naturalness. Ann Arbor: Karoma.

Quirk, Randolph, Sidney Greenbaum, Geoffrey Leech and Jan Svartik. 1985. A
Comprehensive Grammar of the English Language. Harlow: Longman.

Wurzel, Wolfgang U. 1984. Flexionsmorphologie und Natürlichkeit: Ein Beitrag zur
morphologischen Theoriebildung. Berlin: Akademie-Verlag.

K. Aaron Smith is an associate professor of linguistics in the Department of English at Illinois State University. His specialization is language change, particularly focused on verbal systems in English and other Germanic languages.

Format: Hardback
ISBN: 0521884977
ISBN-13: 9780521884976
Prices: U.S. $ 115.00
U.K. £ 60.00