Publishing Partner: Cambridge University Press CUP Extra Publisher Login

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."

The LINGUIST List is dedicated to providing information on language and language analysis, and to providing the discipline of linguistics with the infrastructure necessary to function in the digital world. LINGUIST is a free resource, run by linguistics students and faculty, and supported by your donations. Please support LINGUIST List during the 2017 Fund Drive.

Review of  Constructional Change in English

Reviewer: Stefan Hartmann
Book Title: Constructional Change in English
Book Author: Martin Hilpert
Publisher: Cambridge University Press
Linguistic Field(s): Historical Linguistics
Subject Language(s): English
Issue Number: 25.494

Discuss this Review
Help on Posting

Martin Hilpert’s “Constructional Change in English” presents a constructionist approach to language change. More specifically, Hilpert argues that the notion of constructional change provides a valuable analytical concept in assessing processes of linguistic change. Three corpus-based case studies from the domains of English allomorphy, word-formation, and syntax demonstrate that the idea of constructional change can substantially contribute to a deeper understanding of the diachronic developments in question.

The book is divided into six chapters. The first two chapters introduce the concept of constructional change as well as the corpora and analytical methods used in the study, while Chapters 3 through 5 are dedicated to case studies of English allomorphy, morphology, and syntax, respectively. Chapter 6 then sums up the empirical findings and theoretical considerations and points out some desiderata for further research.

In Chapter 1, the notion of constructional change is defined and some of its implications are outlined. According to Hilpert, “Constructional change selectively seizes a conventionalized form-meaning pair of a language, altering it in terms of its form, its function, any aspect of its frequency, its distribution in the linguistic community, or any combination of these” (p. 16). On this definition, constructional change is not coextensive with either grammaticalization or language change. With regard to grammaticalization, Hilpert points out that the notion of constructional change includes processes of lexicalization, syntactic changes such as the loss of V2 in English, processes within derivational morphology, and processes of frequency change, which all do not instantiate grammaticalization processes. On the other hand, certain grammaticalization processes go beyond constructional change. For example, the emergence of the English modal auxiliaries (e.g. “might”, “could”, “would”) can be seen as a case of paradigm formation, in which a macro-construction in the sense of Traugott (2008) arises not from one single constructional change, but as the sum total of several low-level constructional changes. Delimiting constructional change from language change, Hilpert points out that many instances of language change do not seize individual constructions, but rather apply across constructions. For example, a sound change such as the Great Vowel Shift applies across all word classes and syntactic contexts.

Chapter 2 is dedicated to “Data and methodology”. First, the corpora exploited for the case studies are introduced, each with a brief comment on their size, the represented text types, and their usefulness for the study of constructional change. Then, the statistical methods used to assess the corpora are discussed. Apart from well-established measures of frequency and productivity, Hilpert uses Variability-based Neighbor Clustering (VNC), binary logistic regression, Hierarchical Configural Frequency Analysis (HCFA), and Multidimensional Scaling (MDS). VNC is a bottom-up method to identify stages in diachronic corpora. The VNC algorithm computes the similarity of temporally adjacent data points, using a measure of dispersion such as the standard deviation or the variation coefficient (cf. also Gries & Hilpert 2008, 2012). The two most similar data points are then merged, which is repeated until all data points have been merged. The data produced by the algorithm can then inform a more motivated periodization than would be possible by mere “eyeballing” of a frequency plot. Moreover, VNC can also be used to detect outliers in fine-grained historical data.

Binary logistic regression is a multivariate method to study how speakers choose between two alternatives, e.g. between the “will” and “be going to” futures. While this method is usually applied to analyze synchronic phenomena, Hilpert shows that including time as an additional binary factor into the regression model can shed light on diachronic processes. For example, in his regression model for the two future variants, the factor of time, despite having no main effect in and of itself, interacts significantly with the variable of register. This points to a diachronic leveling process of the register-specific difference between “will” and “be going to”.

Compared to binary logistic regression, HCFA, which is a multivariate extension of the chi-squared test, takes a more exploratory approach. The basic goal of HCFA is to detect similarities between subjects within a population with regard to several variables by performing multiple chi-squared tests. Applied to constructional change, HCFA can determine the prototype structure of a construction and its variants. For example, Hilpert’s analysis of preposition stranding (e.g. “That’s what I was looking *for*”) in the Diachronic Corpus of Present-Day Spoken English (DCPSE) reveals that between 1960 and 1990, preposition stranding spreads into more elevated registers and diversifies syntactically in informal registers.

MDS is another multivariate method to detect mutual similarities between entities within a population. Unlike binary logistic regression and HCFA, however, it reduces the high dimensionality of a multivariate data set to a smaller number of dimensions (typically two) by calculating distances between each possible pair of entities. These distances are then submitted to a scaling algorithm, which transforms the measurements into a set of x- and y-coordinates. Thus, the similarities and dissimilarities between different entities can now be visualized in a two-dimensional plot. Comparing a diachronic series of MDS analyses allows for detecting patterns of change and continuity with regard to a specific construction. For example, Hilpert’s analysis of 44 complement-taking predicates such as “suggest” or “hope” in the Corpus of Historical American English (COHA) between 1860 and 2000 reveals three clusters, which remain relatively stable over time. However, some individual predicates change in their complementation behaviour, and all three verb clusters exhibit a growing preference for “-ing”-clause complements.

Chapter 3 introduces the first case study. The development of English first and second person possessive determiners (“mine/thine” > “my/thy”) is investigated on the basis of data from the Penn and Helsinki Parsed Corpus of Middle English (PPCME) and the Penn and Helsinki Parsed Corpus of Early Modern English (PPCEME) using Variability-based Neighbor Clustering (VNC) and linear mixed-effects modeling. Hilpert finds main effects for the variables of time, phonological environment, stress pattern, priming of the n-variant in the preceding context, formality, and gender. Importantly, however, the variable of grammatical person yields no significant effect. Consequently, Hilpert argues that the change from “mine” to “my” and from “thine” to “thy” should be viewed as one single constructional change, suggesting that speakers formed a constructional generalization over these two linguistic forms. Furthermore, an investigation of the relative frequency distributions of the right-side collocations of the n-variant shows that the prediction that frequent collocates are more resistant to change does not always hold, indicating that “constructional change can follow very idiosyncratic pathways that are subject to multiple interacting factors.” (p. 106)

Chapter 4 analyzes the diachronic development of the word-formation pattern “V-ment” based on the Oxford English Dictionary (OED) and using Baayen’s (e.g. 2009) productivity measures as well as Variability-based Neighbor Clustering (VNC) and Hierarchical Configural Frequency Analysis (HCFA). Drawing on the results of the VNC analysis, Hilpert divides the corpus into five stages. The development of the V-ment construction over time is analyzed by comparing the types yielded by the HCFA for each of the five periods. The overall findings indicate that the developments are not strictly unidirectional, which is why the diachronic change of the V-ment pattern cannot be seen as a case of grammaticalization. The range of possible host classes becomes narrower over time, but this narrowing is not systematic, as the brief success of deadjectival derivatives such as “merriment” shows. Hilpert argues that the concept of constructional change provides a way to come to grips with the idiosyncrasies of these historical developments. Moreover, the development of the V-ment construction, in Hilpert’s view, illustrates the necessity of constructional subschemas (cf. Booij 2010) as a conceptual tool for the study of word-formation processes. He argues that the V-ment construction cannot be regarded as a single word-formation process; instead, different subschemas of the construction rise and fall in productivity.

Chapter 5 deals with constructional change in syntax, illustrated by a case study on concessive parentheticals (e.g. “Power, *although important,* is not everything”). Drawing on data from the TIME corpus of American English and the Corpus of Historical American English (COHA), Hilpert’s analysis addresses two questions: First, are concessive parentheticals derived from full concessive clauses (the reduction hypothesis) or did they emerge in analogy to parenthetical structures with temporal or conditional conjunctions such as “while” and “if” (the analogy hypothesis)? A synchronic comparison of conditional and concessive parentheticals with “if” and “while” in the TIME corpus yields no confirming evidence for the analogy hypothesis and only suggestive evidence in favor of the reduction hypothesis. Second, he asks if concessive parentheticals constitute a “construction family” or if a larger generalization, a “macro-construction”, has emerged over the past two centuries. To address this question, a Multi-dimensional Scaling (MDS) analysis is performed with data from the COHA. More specifically, concessive parentheticals with “although”, “though”, “if”, and “while” are investigated. The macro-construction hypothesis predicts a mutual assimilation of the four conjunctions. The results, however, suggest that the truth lies in between the two hypotheses: While concessive parentheticals with “though” and “although” indeed become more similar in their usage, the changes undergone by parentheticals with “while” and “if” are more idiosyncratic. These empirical findings are relevant for the theoretical question at which level of abstraction constructions can be posited. Hilpert argues that “[i]n order to find ‘a construction’, that is, a generalization that speakers make, it is necessary to find islands of regularity in the variation of the data set” (p. 203).

The right level of abstraction for a construction is also one of the main points addressed in the conclusion (Chapter 6). Furthermore, Hilpert discusses why a constructional perspective is necessary to adequately describe and account for the empirical findings presented throughout the book. Moreover, he points to desiderata for further research, most importantly suggesting an interactional perspective on constructional change at the discourse level and calling for a common analytical vocabulary in Construction Grammar and interactional linguistics.


In recent years, a variety of studies have shown that important insights can be gained from applying a Construction Grammar (CxG) approach to diachronic data (e.g. Bergs & Diewald (eds.) 2008). In a usage-based constructionist perspective, “language is always situated in context, which also implies a connection between linguistic change and language use” (Fried 2013:419). Martin Hilpert’s monograph can be considered an important step towards a better understanding of this connection. Hilpert makes a convincing case for the viability of the notion of constructional change in diachronic linguistics. Far from being just another addition to an already convoluted inventory of technical terms, the concept of constructional change allows for an adequate description and explanation even of highly non-systematic, idiosyncratic processes of change (cf. also Hilpert 2011:69f.). Furthermore, it can help answer one of the key questions of historical linguistics: “What is actually changing? Forms, functions, form–function mappings, rules, and/or exemplars?” (Hruschka et al. 2009:468) Hilpert’s considerations on the relation between constructional change on the one hand and grammaticalization and language change on the other make clear that constructions in the CxG sense are one, but not the only domain of change.

Concerning the methodological approach, this book sets a new benchmark for future corpus-based studies of language change. Most importantly, Hilpert shows how cutting-edge corpus-analytical methods can be applied to refine the key notion of ‘construction’, especially with regard to the level of abstraction at which constructions can be posited. Well-established quantitative measures and statistical methods such as Binary Logistic Regression are used alongside recently developed ones such as Variability-based Neighbor Clustering (cf. Gries & Hilpert 2008, 2012). In all cases, the choice of both the corpora and the analytical methods is well-justified, and the logic behind each method is explained in an intuitively plausible way.

All in all, Hilpert’s book outlines a highly promising approach to the study of linguistic change in a constructionist perspective and is therefore highly recommended to anyone interested in historical linguistics, Construction Grammar, and usage-based approaches to language in general. Moreover, it has the potential to initiate important discussions in the field of (diachronic) Construction Grammar. To be sure, the notion of constructional change will be subject to debate as well as to further refinement in future research. For example, it will be particularly interesting to investigate how constructional change relates to constructionalization, i.e. the appearance of a new conventional form/meaning pairing as a product of a sequence of constructional changes (cf. Trousdale 2013:32).


Baayen, R. Harald. (2009) Corpus Linguistics in Morphology. Morphological Productivity. In: Lüdeling, Anke; Kytö, Merja (eds.): Corpus Linguistics. An International Handbook. Vol. 2. Berlin, New York: De Gruyter (HSK, 29.2), 899-919.

Bergs, Alexander; Diewald, Gabriele (eds.). (2008) Constructions and Language Change. Berlin, New York: De Gruyter (Trends in Linguistics. Studies and Monographs, 194).

Booij, Geert E. (2010) Construction Morphology. Oxford: Oxford University Press.

Fried, Mirjam. (2013) Principles of Constructional Change. In: Hoffmann, Thomas; Trousdale, Graeme (eds.): The Oxford Handbook of Construction Grammar. Oxford: Oxford University Press, 419-437.

Gries, Stefan Th; Hilpert, Martin. (2008). The Identification of Stages in Diachronic Corpora. Variability-Based Neighbor Clustering. In: Corpora 3, 59-81.

Gries, Stefan Th; Hilpert, Martin. (2012). Variability-Based Neighbor Clustering. A Bottom-Up Approach to Periodization in Historical Linguistics. In: Nevalainen, Terttu; Traugott, Elizabeth Closs (eds.): The Oxford Handbook of the History of English. Oxford: Oxford University Press, 134-144.

Hilpert, Martin. (2011). Was ist Konstruktionswandel? In: Lasch, Alexander; Ziem, Alexander (eds.): Konstruktionsgrammatik III. Aktuelle Fragen und Lösungsansätze. Tübingen: Stauffenburg (Stauffenburg Linguistik, 58), 59-75.

Hruschka, Daniel J.; Christiansen, Morten H.; Blythe, Richard A.; Croft, William; Heggarty, Paul; Mufwene, Salikoko S.; Pierrehumbert, Janet B.; Poplack, Shana. (2009). Building Social Cognitive Models of Language Change. In: Trends in Cognitive Sciences 13, 464-469.

Traugott, Elizabeth Closs. (2008). Grammaticalization, Constructions and the Incremental Development of Language. Suggestions from the Development of Degree Modifiers in English. In: Eckhart, Regine; Jäger, Gerhard; Veenstra, Tonjes (eds.): Variation, Selection, Development. Probing the Evolutionary Model of Language Change. Berlin, New York: De Gruyter (Trends in Linguistics. Studies and Monographs, 197), 219-250.

Trousdale, Graeme. (2013). Gradualness in Language Change. A Constructional Perspective. In: Ramat, Anna G.; Mauri, Caterina; Molinelli, Piera (eds.): Synchrony and Diachrony. A Dynamic Interface. Amsterdam, Philadelphia: John Benjamins (Studies in Language Companion Series, 133), 27-42.
Stefan Hartmann is a PhD student in historical linguistics at the University of Mainz, Germany. He is currently conducting a corpus-based study on the diachronic change of German nominalization patterns. Apart from historical linguistics and corpus linguistics, his research interests include Cognitive Linguistics, Construction Grammar, psycholinguistics, sociolinguistics, and language evolution research.

Format: Hardback
ISBN-13: 9781107013483
Pages: 236
Prices: U.S. $ 99.00