LINGUIST List 32.1347

Thu Apr 15 2021

Review: English; Historical Linguistics; Psycholinguistics: Hundt, Mollin, Pfenninger (2020)

Editor for this issue: Jeremy Coburn <>

Date: 23-Sep-2020
From: Pouya Vakili <>
Subject: The Changing English Language
E-mail this message to a friend

Discuss this message

Book announced at

EDITOR: Marianne Hundt
EDITOR: Sandra Mollin
EDITOR: Simone E. Pfenninger
TITLE: The Changing English Language
SUBTITLE: Psycholinguistic Perspectives
PUBLISHER: Cambridge University Press
YEAR: 2020

REVIEWER: Pouya Vakili, Illinois State University


This edited volume has introduced an innovative approach by integrating historical linguistics (which studies language change in communities at some certain points in time) and psycholinguistics (which investigates individual’s cognitive processes) together. These two linguistic subdisciplines have always ignored each other, but Hundt, Mollin and Pfenninger have found “a serious gap in the understanding of processes of linguistic change” (p. 1). While two factors of language change “language-internal and social factors” (p.1) have been studied thoroughly, the last component-- psycholinguistic factors-- needs to be addressed in more detail. The main purpose of this volume is to pave the path to bring historical linguistics and psycholinguistics together, despite their weak connection, to establish an interdisciplinary approach to language change. This aim can’t be achieved unless “experts from the two disciplines [are brought] together in order to explore the potential (and limitations) of an interdisciplinary approach to language change” (p.3). For this purpose, this volume discusses the same cognitive processes from psycholinguistic and historical linguistic viewpoints in each section of the book.

Following this introduction by the editors, the book comes in seven parts--frequency, salience, chunking, priming, analogy, ambiguity, and acquisition--which will be discussed in two separate papers: one from a psycholinguistic perspective and another from historical corpus linguistics perspective.

The first paper in Chapter 2--by Baayen, Tomaschek, Gahl, and Ramscar--talks about frequency and it is entitled “The Ecclesiastes principle in language change”. In this paper, Baayen et al. --after presenting some previous studies--discuss that individuals and societies are exposed to a dramatic increase in knowledge of names, words, collocations, products, etc. They refer to this exponential increase of knowledge as a “suffer” based on a quote “he that increaseth knowledge increaseth sorrow” (Qohelet, Ecclesiastes 1:18). One of the most important results of their study is that the older people grow, the better they master the words in that language (p.31). In addition, older people are more accurate in their articulations, lexical decisions, and collocations. However, they believe that speakers grow slower in various aspects as they grow older, for instance, in learning “nonsense” (here, it refers to learning word pairs that co-occur very rarely like obey-inch). Psychologically, older people show a slower reaction which can be due to “an inevitable process of cognitive decline” (p.37). Baayen et al., however, believe that this slower reaction can be because of the Ecclesiastes principle rather than a decline in cognition since these older people, for their mastery in words, need to search in a larger library. In addition, they state that “Under normal conditions of cognitive and social stimulation, there is no reason to suppose that mental capacities decline with age” (p.48). Baayen et al. conclude that language change is a frequent and ongoing process as individuals grow older and societies develop to become larger.

Hilpert introduces five types of frequency--text frequency, relative frequency, type frequency, burstiness (dispersion), and behavioral profile frequency--in the next chapter. He also points out how these frequencies can be practiced in the cognitive study of language. He maintains that the first three frequencies are well known since they discuss token frequency, tokens of one unit vs. that of another, number of different variants in a corpus in turn while the last two haven’t been addressed sufficiently in historical corpus linguistics. Burstiness measures how (un)evenly the occurrence of a unit is found across the corpus. This can range from single intervals to unpredictable and uneven intervals. Moreover, behavioral profile frequencies compare the frequency of one variant with others. This frequency can reflect that “speakers’ choices between such variants are guided by the morphosyntactic and pragmatic context in which the utterance is made” (p.64). This frequency can indicate the number of options speakers have for alternative linguistic units. Hilpert believes that historical corpus linguistics can help us realize more “possibilities” (p.68) regarding psycholinguistics.

Salience is discussed in Part II, and Nick C. Ellis presents the first paper with the title of “Salience in language usage, learning, and change”. Initially, Ellis provides a definition for salience from a psychological perspective “to refer to the property of a stimulus to stand out from the rest” (p.71). He furthers this definition by proposing that salience can occur in three ways: the physical intensity of the stimulus (like being louder, heavier than the rest), our “resultant knowledge”, cognitive and world experience associations with the stimulus, and the element of surprise and expectations. Accordingly, the first aspect deals with language form, the second discusses language meaning and the third item is associated with linguistic sign in a context, but not consistent with prior language experience. Considering these three aspects of salience, Ellis elaborates on the effects of these parameters in language processing, acquisition, and change. He also maintains that although some potentials can be observed in the meeting of psycholinguistics and historical linguistics, the difficulties, which are due to different methods, theories, scholars, and cognitivists, shouldn’t be overlooked. He concludes that the interdisciplinary study of psycholinguistics and historical linguistics is possible and “something has to come from this” (p.91).

The discussion of salience continues in the next chapter by Traugott asking “How do grammatical items come to have low salience?”. In order to answer this question, Traugott provides an extensive list of features of salience in synchronic linguistics comparing with Ellis’ psychological assumptions. The comparisons indicate that some features overlap, and some others conflict with Ellis’s. Traugott believes that low-salience might have happened in phonology, but due to the unavailability of evidence, a certain conclusion can’t be drawn. In addition, low salience certainly facilitates change in a grammatical item such as BE going to, shall and will in English, and-r in Latin. Traugott adds that psycholinguistics and historical linguistics haven’t yet achieved a ubiquitous description of salience (to work in both disciplines); therefore, she suggests that the psychological salience suggested by Ellis needs to be redefined. Since “low salience is an important enabling factor in morphosyntactic change” (p.108), the new definition needs to consider individual, social and general semantic/cognitive aspects separately.

Chunking is the subject of discussion in Part III, and it starts off by Ellis’ paper entitled “Chunking in language usage, learning, and change: I don’t know”. Chunks are defined as sequences of stimuli (like sequences of sounds, letters, morphemes, words, or phrases). Speakers’ prior knowledge, experience, and exposure are assumed to be the most important determining factors in realizing and distinguishing these chunks as units. Therefore, speakers’ judgments are beyond the semantics and formality of these units. Ellis states that chunking is rational since it relies on the rational analysis (Anderson, 1990) and its purpose is to “answer why human cognition is the way it is” (120). That might be the reason why psycholinguistics has considered everything in chunks (phonotactic, lexical, orthographic, morphological, collocation, phrasal, sentence processing, and hieratical chunks). After discussing connectionism and statistical language learning, and delving into construction grammar and language change, Ellis recapitulates Zipf’s laws and principles regarding language chunks in language change. He finds Zipf’s laws and principles prominent since they elaborate on language use and change using a simple formula which is psychologically and computationally practical at any level of individual, community, language acquisition, language use, and change. Ellis admires Zipf’s laws and principles since these ideas had been introduced much before the emergence of historical linguistics or psycholinguistics, and even before the invention of computers.

Chapter Seven by Bybee and Moder is a response to Ellis’ paper on the issue of chunking. The focus of this chapter is to examine beg the question as a case study diachronically. Bybee and Moder disregard Ellis’ proposition that views chunking as a favorable unit for speakers. They claim that chunking includes “processes by which chunks acquire meaning from the context” (p.148). Additionally, they reject Zipf’s laws and principles, brought up in the previous chapter, stating “teetering dangerously on the verge of teleology” (p. 152) because they assume Zipf’s theory is “goal-oriented” and this theory looks at language as a whole (p.152). Bybee and Moder believe that language changes because this is what speakers want and this is one of the qualities of language. They believe that no goal or plan should be presupposed in the language since the primary purpose of language is to communicate “not to create new grammatical theories” (p.152). In Bybee and Moder’s scheme, chunking is a process that happens gradually over time, and different chunks are created since people feel the need to have them. Unlike Ellis’ claim, they believe that “humans automatically chunk all of their experience, not because it creates more efficient communication (p.152-3). Bybee and Moder trace the origins of beg the question by examining different databases and gathering around 1480 tokens. Citing Liberman (2010), they find out that this collocation is the result of “misleading translations” (p.156) and its history goes back to the Aristotelian period in which it was used to signal logical fallacy coming from Latin. They provide an exhaustive list of examples in which this collocation has been used and maintain that it was in the late 17thcentury that this expression found a fixed position with less compositional meaning. Shifting to “faulty reasoning” in the next century, and then “compel to ask” in the mid-20thcentury, Bybee and Moder conclude that context is the most essential leading factor in the analyzability and compositionality of any chunks.

Section IV in the book is dedicated to Priming with the first paper by Pickering and Garrod entitled “Priming and language change”. They define priming as “a largely non-conscious or automatic tendency to repeat what one has comprehended or produced” (p. 173) which can affect both language use and development. Pickering and Garrod propose a one-week life span for priming compared to the continual life span of language change with having effects over decades or even centuries. Pickering and Garrod discuss alignment in the first part of this chapter and structural priming comes later. Alignment has been defined as a “central goal of conversation” (p.173) which can result in linguistic changes in an individual. The effects are not so long-lasting in that if the individual is set in another conversation, new alignments might be created “I immediately copy your form and then remember the form I have just used” (p. 179). Alignment can lead to routinization which might last longer than alignment itself, but if these routines persist much longer than usual, they can result in language change. Structure priming happens which interlocutors “repeat the syntactic structure of utterances that they have just produced (Bock, 1986) or comprehended (Bock et al. 2007)” (p. 177). Pickering and Garrod believe that structural priming is more likely to happen in conversations between native speakers or well-learned second-language speakers. Not surprisingly, since priming is directly connected to imitation, it might lead to ungrammaticality because the speakers who have been exposed to an ungrammatical structure tend to produce the same structure without thinking about that. It is suggested that more out of the laboratory and “in situ” studies over different generations should be conducted in order to be able to propose a solid argument for this purpose. Pickering and Garrod conclude that priming, alignment, and routinization are possible language change reasons in which adult speakers of L1 and L2 may contribute.

Section IV continues with a paper by Christian Mair entitled “From Priming and Processing to Frequency Effects and Grammaticalization? Contracted Semi-Modals in Present-Day English”. In this chapter, Mair tries to approach priming from the historical linguistic perspective. He admits priming, alignment, and routinization processes suggested by Pickering and Garrod, and suggest these processes need to be investigated more thoroughly so that they can be connected to historical linguistics equally. Unlike Pickering and Garrod who believed that language change can happen at an individual level, Mair, citing Weinreich, Labov and Herzog, states that “the site of language change is the community and not the individual” (p.193). Therefore, cross-validate studies on recent language changes from corpus analyses are required to investigate the issues of priming and alignment. Mair points out that there are some weaknesses and gaps in historical linguistics, and psycholinguistics, which need to be reconsidered so that it can go alongside psycholinguistics to form historical psycholinguistics. In the same vein, he concludes that “the psycholinguistic perspective is needed alongside the sociolinguistic and discourse-pragmatic ones in order to develop a truly comprehensive, usage-based model” (p.209).

Analogy is the topic of part V, and the first paper is presented by Behrens with the title of “The Role of Analogy in Language Processing and Acquisition”. Using Gentner’s (1983) definition for analogy, she believes “Analogy is a domain-general form of structure mapping between a source and target” (p.216). Behrens identifies analogy as a domain-general mechanism for categorization and category formation (p.215). This categorization, as she draws, can be of two similarities either perceptual or rational. She states that despite perceptual and relational similarities in entities to shed light on the mechanism of analogy, entities can be problematic due to their unpredictability. Behrens believes that analogy needs to be used practically in language acquisition since it explains about language change and development from infant to a competent adult speaker while historical linguistics has used this term more often to discuss language change in a community from the past to the present. Behrens supports her analogy theory by the overgeneralization of regular -ed past tense marker which can lead to some errors such as goed, instead of went, or pluralization such as foots instead of feet. Citing Gentner and Smith (2012:131), Behrens introduces three processes of analogical reasoning; Retrieval, Mapping, and Evaluation (p.217). These processes share similarities with language change and development in language acquisition and historical linguistics. Behrens argues that psycholinguistics should also benefit from analogical reasoning for their language acquisition schemes. She also adds that analogy can be practiced in frequency and salience since they have complex interactions with each other. In conclusion, Behrens encourages to collect experimental and observational evidence from language change and acquisition to form an empirical basis.

“The role of analogy in language change: supporting construction” is the title of chapter 11 by De Smet and Fischer. As the title shows, this chapter comes into two parts; in terms of the role of analogy in language change, they state that “if analogy is to be properly understood, its operation must be seen against the background of complex constructional networks capturing the myriad relations between individual constructions”(p.242). Additionally, in terms of supporting construction, they, citing Abbot-Smith and Behrens (2006:1,019), propose the reason why some constructions are acquired earlier and faster than others. De Smet and Fischer believe that since some subparts in a construction are already mastered, these subparts can be experienced in known constructions, and when more known constructions are practiced in these subparts, new construction will be acquired faster and earlier. This is a cyclic mechanism that can happen in language acquisition and likewise in language change. They maintain that language change requires “analogical extension” which can shed light “on one or more already existent patterns” (p.243). They also use grammaticalization theory to discuss have to and as good as, and they conclude that “the timing and context of analogical change is not random” (p.263). This is a counterargument against the critics who consider analogy too unconstrained and unpredictable.

Ambiguity is the topic of part VI, and the first paper with the title of “Syntactic ambiguity in real-time language processing and diachronic change” is presented by Claudia Felser. Not only does she consider ambiguity in language a good thing, but she also agrees with Piantadosi et al. (2012) that “ambiguity increases a linguistic system’s communicative efficiency” (p.272). Felser, then, discusses how psycholinguistics can prevent misanalysis of ambiguity by exemplifying research on morphosyntactic ambiguity. As she states these studies have been conducted on groups in laboratory environments to investigate “how language processing system deals with ambiguous input” (p.272). After that she investigates individual speakers’ analysis of ambiguous structures and concludes that this reanalysis can affect a community’s language system and they, consequently, may lead to language or grammar change. In this chapter, Felser’s main focus was ambiguity in syntax, and she calls for more studies on neoanalysis in semantic and pragmatic factors.

David Denison discusses “Ambiguity and vagueness in historical change” in chapter 13. His focus is on words rather than syntactic structures. Initially, Denison differentiates ambiguity from vagueness in that ambiguity is defined as two distinct interpretations of a statement with typically one of them is intended by the speaker/writer, while vagueness “is where a linguistic analysis is in some relevant respect underdetermined at least for addresses or readers …..” (p.293). Denison believes that ambiguity is a primary characteristic of words and it is a semantic phenomenon which is independent of context. For example, the word bank has several meanings, and these meanings can create ambiguity. However, vagueness is a pragmatic phenomenon that is context-based, and the addressees/readers need to draw their interpretations from the information provided to them in the context. Denison, examining corpus data and dictionaries over time, points out that vagueness, not ambiguity, is a “driver of change.” (p.294) In addition, he states that other factors might be involved in determining how these changes should proceed. In conclusion, Denison sees ambiguity as more of an exterior phenomenon and the one that can be the result rather than the cause of the change.

Acquisition and transmission are discussed in Part VII and the first paper is delivered by Lieven with the title of “Developing language from usage: Explaining errors”. In this chapter, Lieven has focused on various types of errors in the process of language acquisition. She finds that Universal Grammar (UG) model centers on “the problems of mapping between UG and the features of the language that children are learning” (p.321). However, the Usage-based (UB) approach focuses on “interactions between learning processes and features of the input” (p.321). Therefore, when acquiring a language, children make systematic errors which can be the result of learning strings, relative frequency from the input, and blending of two schemas. She points out that these errors are temporary and as the child grows older, these errors vanish so that it can’t be claimed that children are the cause of language change. However, like many other linguists from wide-scheme theoretical approaches, Lieven claims that adolescents, adults and second language learners can be identified as one of the sources of language change.

Lopez-Couso presents Chapter 15, entitled “Transferring insights from child language acquisition to diachronic change (and vice versa). Following Lieven, she also doubts the “validity of a child-centered theory of language change” (p.332). In order to discard this validity, she examines the parallels between ontogeny and phylogeny, and between diachronic and acquisitional developments. She figures out that although various evidence can be spotted for language acquisition and change parallels, these parallels are alike, but not identical. She also indicates that these parallels can be “illusory” or “genuine” (p.338). The former means the principles which lead to acquisition are totally different from the ones that govern language change, and they might accidentally produce similar results, but they are not the same. The latter is defined as the sense of sharing the same psychological and cognitive mechanisms. She studies different parallels in various domains such as irregularity in morphology to become regular, epistemic and deontic meanings, different meanings of present perfect, locative there to existential there, and development of going to-pattern for future applications. In her conclusion, Lopez suggests that in order to understand certain language changes, language acquisition needs to be observed more meticulously and the other way around since the parallels between ontogenetic and diachronic developments and language acquisition and change are to some extent related.


The present volume has been clearly successful in integrating historical linguistics and psycholinguistics and examining their interrelated components in influencing language change. The book clearly shows the junctions where these two disciplines can meet in order to provide an interdisciplinary research avenue for both discipline experts. In fact, the paired papers on a shared topic provides solid evidence that these two disciplines, despite some controversies, can work well together.

In general, considering the topics discussed in this edited volume, it can be pointed out that in terms of frequency and analogy, these two disciplines share a lot together in that you can find some interdisciplinary research on these two topics. In terms of chunking, priming, and acquisition, more studies and research should be conducted in order to find some more common grounds. However, salience and ambiguity require more research to be more compatible with both disciplines.

Each chapter in this book provides many research ideas for different scholars in any linguistics field like corpus linguistics, psycholinguistics, TESOL, etc. both theoretically and empirically. These research ideas can help flourish these two disciplines much better and establish a fully interdisciplinary field.


I am a PhD student in TESOL and Linguistics at Illinois State University. My research interests include syntax, grammaticalization, historical linguistics and corpus linguistics.

Page Updated: 15-Apr-2021