27.4343, Review: English; Socioling; Text/Corpus Ling: Grieve (2016)

[The LINGUIST List via LINGUIST]

LINGUIST List: Vol-27-4343. Thu Oct 27 2016. ISSN: 1069 - 4875.

Subject: 27.4343, Review: English; Socioling; Text/Corpus Ling: Grieve (2016)

Moderators: linguist at linguistlist.org (Damir Cavar, Malgorzata E. Cavar)
Reviews: reviews at linguistlist.org (Anthony Aristar, Helen Aristar-Dry,
                                   Robert Coté, Michael Czerniakowski)
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: Clare Harshey <clare at linguistlist.org>
================================================================


Date: Thu, 27 Oct 2016 13:56:57
From: Sanna Hillberg [sanna.hillberg at uef.fi]
Subject: Regional Variation in Written American English

 
Discuss this message:
http://linguistlist.org/pubs/reviews/get-review.cfm?subid=36182860


Book announced at http://linguistlist.org/issues/27/27-633.html

AUTHOR: Jack  Grieve
TITLE: Regional Variation in Written American English
PUBLISHER: Cambridge University Press
YEAR: 2016

REVIEWER: Sanna Hillberg, University of Eastern Finland

Reviews Editor: Robert A Cote

SUMMARY

This book discusses regional variation in American English (AmE) syntax, which
rather surprisingly is an under-researched field in the English linguistics.
The database of this corpus-based study consists of letters to editor, and it
is compiled by the author, Jack Grieve. He investigates alternation variables
that may be used synonymously and applies a multivariate analysis on the data.
According to Grieve (2016: 36–9), alternation variables investigated here have
been selected on the basis of them being interchangeable in discourse without
changing the meaning they convey. In terms of grammatical features, for
example, that, who and zero may denote a human antecedent in a relative
clause, and all these variants may be used in the same context. Alternation
variables may encode, for example, a user’s regional background as well as
other types of social and situational information. Each particular grammatical
feature included in this study has at least two alternations, for example,
pronouns everyone and everybody.

Chapter 1 “Introduction” outlines the history of American dialectology
starting from Hans Kurath, Marcus L. Hansen, Bernard Bloch, and Julia Bloch’s
(1939) “Linguistic Atlas of New England” to present. Surprisingly, there is
only one large-scale survey on AmE grammar prior to Grieve, that of E. Bagby
Atwood (1953). As a result, the lack of previous research complicates the
identification of alternation variables for this study (Grieve 2016: 39).
Despite the obvious lack of research on AmE grammatical features, there are
several investigations on AmE regional variation in phonology as well as in
social variation in pronunciation. Another point where the current study
differs from mainstream dialectal studies is the type of data used. Whereas
material in dialectal studies typically consists of elicited spoken data, e.g.
questionnaires and interviews, Grieve utilizes written data collected from
newspapers – material so far alien to AmE dialectology (see also Chapter 2).
At the end of Chapter 1, Grieve states his research questions, briefly
describes the main findings of the study and introduces forthcoming chapters.

Chapter 2 “Corpus” first explains differences between corpus-based and
traditional dialectology. Grieve claims that the former may be more economic
and simple, especially in terms of data collection. For example, data may be
collected without lengthy visits to faraway places to interview informants. In
addition, corpus-based methods allow researchers to gather larger databases
than usually collected by elicitation, which again leads to more valid results
and allows examination of features that would not possibly crop up in a small
database. In this chapter, Grieve explains the design, compilation, and
dimensions of his corpus and justifies the genre selected for the study, i.e.
letters to editor. In addition, he discusses problems regarding newspaper
language as a research material. 

In the beginning of Chapter 3 “Grammatical analysis”, Grieve introduces the
investigated features: 15 grammatical categories, 135 grammatical alternation
variables and their 295 variants. The grammatical categories are the
following: pronouns, relative pronouns, determiners, adjectives, nouns,
prepositions, particles, subordinators, coordinators, verbs, modals,
infinitives, not, adverbs and adverbials. Grieve explains the methods of
retrieving, selecting, measuring, and mapping the variables. The latter part
of this chapter is dedicated to the illustrative alternation maps of all
investigated variants. These maps show how the use of each variant is spread
across the United States. In addition, it has informative tables which provide
examples of the use of each alternation variant as well as their occurrence in
the corpus and their distribution. 

Chapter 4 “Spatial analysis” provides detailed information on spatial
analysis, which is rarely used in dialect studies, and applied in this
investigation. First, however, the chapter briefly compares ‘traditional’
standard approaches to dialectology and discusses, for example, the effect of
social background and temporal variation in linguistic research. Grieve also
discusses the points that in his view are the major limitations in traditional
dialectology, i.e. drawing isoglosses on dialect maps based on small sample
sizes and possible researcher bias. 

After this, Grieve explains local autocorrelation analysis, the statistical
method used in this study, in detail. According to Grieve (2016: 100), it “is
based on a standardized and statistically grounded procedure” and “enables the
identification of underlying patterns of spatial clustering in the values of a
linguistic variable”. He argues that the application of this method along with
other statistical tools enables replication of results, which is not easily
achieved by methods used in traditional dialectal studies. 

Grieve also explains how the linguistic data matrix is formed in this analysis
as well as the major concepts related to spatial analysis (spatial dependence,
spatial stationarity vs. non-stationarity, the analysis of spatial
autocorrelation, and statistics related to its measuring). He introduces the
local spatial autocorrelation analysis, which he applied with the local
Getis-Ord Gi statistic used for identifying patterns of spatial clustering,
especially non-stationary patterns. Next, he explains how the results are
turned into local-spatial autocorrelation maps that show clustering of
variants and provides detailed description on how spatial weights matrix may
be applied in investigating alternation variants and turned into maps that
show clustering. Again, there are illustrative maps that indicate comparison
of spatial weights matrices for selected features. 

At the end of this chapter, Grieve provides local spatial autocorrelation maps
for all the variants of the study. The maps show locations that have positive
z-scores, that is, locations that have relatively high percentages of the
variant. In a fairly short discussion of the results, Grieve points out two
major findings of the investigation. First, “the majority of variants show
some degree of spatial clustering”. Secondly, “these patterns are diverse and
complex” (2016: 122). Despite the variation being somewhat weak, the findings
show that syntax in letters to editors do not function in uniform across the
United States. The last pages of this chapter are dedicated to the discussion
of the replicability and consistency of the statistical method used here in
comparison to other possible methods.  

Chapter 5 “Multivariate analysis” concentrates on the statistical methods
applied in this work: multivariate analysis, factor analysis, and fuzzy
cluster analysis. The chapter begins by explaining the major differences
between dialectometry and multivariate analysis, then moves on to provide
detailed description on how the three types of statistical analyses are
applied in this study in order to identify patterning of syntactic features
and major AmE dialect regions. 

Grieve (2016: 165) explains the reason why three factors have been selected as
the optimal number of factors to be extracted from the data matrix. Again,
this chapter has illustrative figures and tables. For example, Table 5.2 shows
loadings for factors 1–3 as well as variable communalities for the selected 83
variables by 240 locations. Grieve also provides explanations as to what the
maps indicate and which regions the factors 1–3 contrast. The factors seem to
pattern on the maps so that they correspond to major American cultural
patterns. Towards the end of this chapter, scatterplots indicate consistent
regional clustering. In addition, fuzzy cluster analysis identifies five major
dialect regions, which “clearly correspond to important American cultural
regions” and transition areas (2016: 184). 

Chapter 6 “Sources of regional linguistic variation” discusses the reasons
behind regional variation detected by various statistical analyses. The
present findings are compared to those in COCA corpus. Both language internal
and external reasons are discussed as explanations for variation. As for
language internal variation, results indicate that regional variation occurs
in relation to the use of formal vs. informal variants even though the
differences across regions are not particularly strong. External explanations
to regional variation are also related to formality in language use. This
investigation shows that settlement patterns, which earlier studies have
typically used as an explanation for regional differences in AmE, are not a
sole explanation for variation. In addition, physical and cultural boundaries
affect language use across the United States. 

Factor 1 indicates that patterns of population density play their part in
regional syntactic variation. More formal forms of the variants are used in
the eastern United States, which is more populated than the western parts of
the country. Factor 2, on the other hand, “identifies a difference between the
South Central States and the rest of the United States” (p. 201). Cultural,
geographic and population density patterns between these regions are used as
explanations for variation. Factor 3 identifies a difference between the
Midwestern areas vs. the rest of the United States (especially the Southeast).
Various external factors, i.e. physical, historical, cultural, and
administrative factors explain the differences. In order to support his claim
that these external factors affect linguistic variation, Grieve presents
convincing maps of elevation, affluence, population density, 2008 presidential
election (democrats vs. republicans), ancestry, and Census bureau regions.
Based on the findings of the study, Grieve proposes the existence of five
dialect areas in the United States (i.e. West, Midwest, South-Central,
Northeast, and Southeast) instead of the two or three dialect areas identified
by previous studies. For example, Craig M. Carver (1987) proposes a two-way
division (the North and the South), whereas Hans Kurath (1949) identifies
three major dialect regions (the North, the Midland, and the South). The end
of this chapter provides a brief comparison of internal and external
explanations to variation, comparison to previous findings on regional
variation in AmE, and predictions on the development of future American
dialect regions. 
 
Chapter 7 “Conclusion” ends the book by restating the aims, material, methods,
and major findings. Appendices A to C introduce in detail the corpus as well
as the alternation variables, their search patterns, and descriptive
statistics. Appendices D and E provide the results of spatial and multivariate
analyses.

EVALUATION

Grieve’s study on AmE syntax is timely, long overdue, innovative, and
thorough. He takes a fresh approach to investigating dialectology, both in
terms of material and methods. The book is a well-structured and coherent
description of modern AmE grammar. It introduces innovative statistical
approach to the study of dialectal variation, and concentrates on written
language instead of spoken language, which is typically the object of study in
dialectology.

The book makes use of highly illustrative maps, which help the reader to
understand the rather complex statistical results. In addition, figures and
tables provide valuable information about the features investigated.

The study at hand would have benefited from comparisons to other genres in AmE
in order to confirm that the detected differences exist across speech and
writing. However, considering the amount of work the investigation of all the
135 alternation variables required, it is understandable that their
examination in other genres has been left for future research. 

The two unquestionable merits of this work are one, it is the first
large-scale study on AmE grammar, and two, it has opened a path for a new way
of investigating AmE  – or any other language or its variety. Those who
investigate regional linguistic variation will greatly benefit from the
statistical methods introduced and applied in this work.

At the end of this book, Grieve argues that “communication will always be
constrained by physical distance… leading to greater regional diversity in
American English” (p. 218). It will be interesting to see whether this
argument holds true. Our world is constantly shrinking with new media and easy
and affordable travelling. As a result, physical boundaries do not limit
communication as they did earlier. Only future will tell whether regional
dialectal variation will thrive or level out. 

This book is excellent reading for anyone who is interested and investigates
grammatical variation in English(es). It does not only provide new angles to
the study of a very traditional field such as dialectology, but it also
provides excellent material for comparisons on the use and spread of syntactic
features with other Englishes around the globe. This book should also motivate
researchers to investigate syntactic variation and written language use in AmE
more closely.


ABOUT THE REVIEWER

Sanna Hillberg currently works as a post-doctoral researcher at the University
of Eastern Finland. Her main research interests include Scottish English,
Scots, syntax, sociolinguistics, variation and change.





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

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

        Thank you very much for your support of LINGUIST!
 


----------------------------------------------------------
LINGUIST List: Vol-27-4343	
----------------------------------------------------------