Publishing Partner: Cambridge University Press CUP Extra Publisher Login

New from Cambridge University Press!


Revitalizing Endangered Languages

Edited by Justyna Olko & Julia Sallabank

Revitalizing Endangered Languages "This guidebook provides ideas and strategies, as well as some background, to help with the effective revitalization of endangered languages. It covers a broad scope of themes including effective planning, benefits, wellbeing, economic aspects, attitudes and ideologies."

New from Wiley!


We Have a New Site!

With the help of your donations we have been making good progress on designing and launching our new website! Check it out at!
***We are still in our beta stages for the new site--if you have any feedback, be sure to let us know at***

Review of  The Mental Corpus

Reviewer: Ross Deans McLachlan
Book Title: The Mental Corpus
Book Author: John R. Taylor
Publisher: Oxford University Press
Linguistic Field(s): General Linguistics
Philosophy of Language
Cognitive Science
Issue Number: 27.1482

Discuss this Review
Help on Posting
Reviews Editor: Robert Arthur Cote


In this book, “The Mental Corpus”, John R. Taylor is concerned with a single, overarching question – what does it mean to know a language? In contrast to conventional accounts, Taylor argues across thirteen chapters that linguistic knowledge is a conglomeration of the individual’s experiences with the utterances of a language. Taken together, this body of usage events is the eponymous ‘mental corpus’ of the book. Drawing on data taken primarily from the British National Corpus (BNC) and the Corpus of Contemporary American English (COCA), Taylor presents a sustained argument for a radical reconceptualization of the nature of language.

The first chapter, ‘Conceptualizing Language’, introduces the generative model of linguistic knowledge (cf. Chomsky 1986, 1995) which Taylor aims to replace. Specifically, Taylor frames the discussion as one between external language (E-language) and internal language (I-language). This distinction leads generative approaches to prioritise the I-language as the proper subject of linguistic enquiry at the expense of E-language (Isac & Reiss 2013). In contrast to this, Taylor instead argues that the relation between E-language and I-language is instead a dialectic one. On one hand, E-language is the output the I-language of individual speakers; on the other, a speaker’s I-language is a result of their exposure to the E-language. As I-language can only be inferred, so Taylor argues that the primary source of evidence must be found in E-language. Taylor rounds off this opening chapter a discussion on what constitutes representativeness in corpora, including an interesting section on the World Wide Web as a ‘fabulous linguists playground’ (Kilgarriff & Grefenstette 2003: 345).

Chapter 2, ‘The Dictionary and the Grammar Book’, continues with a rejection of the generative model of linguistic knowledge or what Taylor perhaps somewhat flippantly refers to as the ‘dictionary-plus-grammar-book model’. He notes that this bifurcated model of linguistic knowledge has dominated the research agenda of academic linguistics for the past half century. As Taylor sees it, the widespread popularity of this approach has been as a result of the way it accounts for linguistic creativity, offering a certain undeniable and intuitive appeal. However, there would seem to be a number of flaws in this approach, with the generative model unable to account for a number of linguistic phenomena. For instance, drawing on attested corpus data, Taylor shows the regular occurrence of ‘explain’ within a ditransitive frame which would seem to violate certain principles of grammaticality. He argues that by labelling such occurrences as ‘idiomatic’ and shifting them to the periphery, the generative model undermines its own validity.

In Chapter 3, ‘Words and their Behaviour’, Taylor turns his attention to the distribution of words in actual language use. In this case, the argument is that the simple sorting of lexical items into lexical categories is in fact fraught with inadequacies. From the corpus data he employs, Taylor argues for example that the various usages of ‘much’ and ‘fun’ are not exhausted by their respective categorizations as quantifier and (singular) mass noun. Instead, the argument put forward is that a word’s syntactic distribution may be entirely sui generis. In this respect, this chapter presents important corpus based evidence in favour of a position such as Croft’s Radical Construction Grammar (2001).

Returning to the earlier theme of idiomatic language, Chapter 4, ‘Idioms’, and Chapter 5, ‘Speaking idiomatically’, treat the phenomenon together and at length. It is across these two chapters that Taylor’s use of corpus data is at its strongest and most convincing. Evidence of the pervasiveness of ‘idiomatic’ expressions is shown, with examples falling under one of four categories: semantic, syntactic, lexical, and phrasal. The conclusion we are led to accept in Chapter 4, therefore, is that a great deal of everyday language usage could be classed as idiomatic in one way or another. In this respect, Taylor argues that the generative model ‘undergenerates’, failing to predict a non-trivial amount of language used by proficient speakers. Similarly, the model often ‘overgenerates’, resulting in perfectly grammatical sentences which sound unnatural to speakers of the language. In contrast to this, Taylor proposes an alternative conceptualization. On this account, we understand the idiomatic in terms of conformity with usage norms. These usage norms are contextually situated (Halliday & Matthieson 2014) and statistical, rather than categorical, in nature (Firth 1957). One of Taylor’s main contentions is that, although it is impossible to ‘teach’ these usage norms fully, they have clearly been learned by proficient native speakers in constructing their own mental corpus.

Chapter 6 goes by the title ‘Constructions’. Here, Taylor expounds a more explicit account of the notion of ‘construction’, with a particular emphasis afforded to Langacker’s Cognitive Grammar (1987, 1991, 2008). Building on the slightly different notions of constructions as internally complex entities and constructions as form-meaning pairings, Taylor presents a third view of constructions as ‘units’. A unit, in this respect, refers to ‘whatever a speaker of a language has specifically learned’ (p.126). While such a broad definition of constructions may seem to reduce the descriptive relevance of the concept, Taylor presents a robust case against rule-based approaches and the autonomy of syntax hypothesis. To quote his provocative claim – ‘It’s constructions all the way up’ (p.145).

In Chapter 7, ‘Frequency’, unpacks some earlier claims regarding the statistical nature of usage norms. Specifically, this chapter address the question of where exactly frequency information belongs in a descriptive theory of language. As in earlier chapters, this discussion is framed by Chomsky’s position, which is that frequencies are part of E-language and therefore irrelevant to linguistic analyses. Taylor works through a wide range of psycholinguistic research, showing the relevance of frequency phenomena in relation to verb complements, word frequency, collocations, phonology, ambiguity resolution and ‘garden path sentences’, and productivity of schematic generalizations. Through a detailed and sophisticated review of this research, Taylor ultimately concludes that frequency effects are not merely a by-product of the E-language. Rather, these effects are an inherently central aspect of a person’s linguistic knowledge.

Chapter 8 expands this discussion, arguing in favour of the chapter heading, ‘Skewed frequencies as a design feature of language’. Focusing here on markedness and categorization, Taylor argues that frequency effects are fundamentally responsible for each of these phenomena. In this way, his approach touches on the nature of prototypicality (Rosch 1975). However, this leads to somewhat unsatisfying conclusion. On this account, one would assume that non-prototypical units with a low-frequency may not be learned by new generations of learners, ultimately resulting in less skewed frequency until a stable period of equilibrium. However, this is not the case of any language. Therefore, perhaps controversially, Taylor proposes that skewed frequencies are an absolutely central feature of human languages.

Chapter 9, ‘Learning from Input’, builds on this claim. Arguing from the perspective of phoneme acquisition, Taylor holds that this in-built skewing of frequency has a functional explanation. Focusing primarily on phoneme acquisition, the assertion in this chapter is that ordinary human development and learning requires this statistical disparity. Making further reference to a body of experimental research into ‘recency effects’ (syntactic priming), this chapter seems to present strong empirical evidence in support of Taylor’s claim regarding the dialectic relationship between E-language and I-language.

Chapter 10, ‘Polysemy’ represents a change of tack, shifting the focus to the nature of linguistic forms with two or more related meanings. Despite being a central feature of cognitive linguistic theory, Taylor argues that this definition of polysemy proves to be problematic. Having read through the previous chapters, the reader is in the position to anticipate the main thrust of the argument. Put simply, words do not represent discrete chunks of form-meaning associations. Instead, words should be thought of in terms of their ‘contextual profile’. Alongside their conventional conceptual content and referential value, this also covers the kinds of items with which they collocate, the constructions in which they feature, and the types of text in which they occur. Therefore, knowing a word involves knowing the contexts in which that particular word can occur. Intriguingly, the word ‘pragmatics’ does not seem to occur anywhere in this discussion.

In Chapter 11, ‘Creativity and Innovation’, Taylor returns to an earlier argument put forward by generativists as unassailable. In this chapter, the individual capacity for seemingly unbounded linguistic creativity is contrasted with the nature of linguistic innovation. Taylor concedes that the generativist position is well-founded in the case of creativity, which remains strictly within the rules of a language system. In contrast, innovation involves going outside of and beyond that system of rules working on a lexicon. The position developed in this chapter is that there are problems with the generativist position in relation to the distinction between creativity and innovation. Taylor deftly returns to earlier discussions of idioms, illustrating that in the case of ‘all over’, there are clear features of ‘idiomatic creativity’, an oxymoron according to the orthodox accounts. There is therefore no clear, logical relation between the existence of linguistic creativity and innovation and the necessity of a generative model of linguistic knowledge.

Not content to stop there, Chapter 12, presents an alternative approach to linguistic creativity and innovation which complements the rest of the book. This alternative is ‘Blending’ which forms the title and focus of this chapter. Taylor draws on Fauconnier & Turner’s (2002) highly influential outline of Conceptual Blending and Fauconnier’s (1994, 1997) related research into Mental Spaces Theory. A full account of the principles behind mental spaces and blending would not be possible in this review. However, whereas Fauconnier and Turner focus primarily on the nature of ‘conceptual’ blending, Taylor here makes a persuasive case for word blending and phrasal blending as a motivating factor behind linguistic creativity and innovation. Most interesting is the discussion of phrasal (or constructional) blending with a number of corpus examples such as ‘keeping an eye out’, ‘ever since I can remember’, ‘time and (time) again’, and a call-back to the earlier ditransitive ‘explain me this’. Taylor maintains that this approach accounts for the same phenomena as the traditional generativist approach, while simultaneously explaining those features that had been relegated or deemed marginal.

Finally, Chapter 13 called ‘The mental corpus’, summarizes the main arguments advanced throughout the book. In this short concluding section, Taylor brings together the individual threads of his monograph: a usage-based approach to language; commitment to corpus data and experimental research; the centrality of idiomatic language; the importance of frequency; and the dialectic between E-language and I-language.


Building on Taylor’s earlier monographs on cognitive linguistics (Taylor 1996, 2002, 2003 [1989]), this book represents a formidable contribution to the literature. Taylor’s writing is lively and accessible and his arguments are well supported by an expanding body of cognitive and corpus approaches to linguistics. Marrying these approaches together so effectively hopefully represents new steps in the maturation of these disciplines.

Perhaps the most striking thing about this book is in the way that it builds upon and adapts the well-known metaphor of ‘the mind is a machine’. What emerges from the text is the innovative, blended metaphor of ‘linguistic knowledge as searchable corpus’. This idea of (linguistic) knowledge as a sort of ‘hypertext’ is something which could potentially open up new avenues of research and enquiry.

Finally, Taylor’s style of presentation makes this text suitable for a large body of readers. It will appeal to cognitive and corpus linguists alike, as well as students and lay readers interested in contemporary linguistic theory.


Chomsky, N. (1986). Knowledge of Language. Its Nature, Origin, and Use. New York: Praeger.

Chomsky, N. (1995). The Minimalist Program. Cambridge, MA: MIT Press.

Croft, W. (2001). Radical Construction Grammar: Syntactic Theory in Typological Perspective. Oxford: Oxford University Press.

Fauconnier, G. (1994). Mental Spaces: Aspects of Meaning Construction in Natural Language. Cambridge: Cambridge University Press.

Fauconnier, G. (1997). Mappings in Thought and Language. Cambridge: Cambridge University Press.

Fauconnier, G & Turner, M. 2002. The Way We Think. Conceptual Blending and the Mind’s Hidden Complexities. New York: Basic Books.

Firth, J. R. (1957). Papers in Linguistics 1934–1951. London: Oxford University Press.

Halliday, M.A.K. & Mattiessen, C. (2014). Halliday's Introduction to Functional Grammar. London: Routledge.

Isac, D. & Reiss, C. (2013). I-language: An Introduction to Linguistics as Cognitive Science. Oxford: Oxford University Press.

Kilgarriff, A. & Grefenstette, G. (2003). “Introduction to the special issue on the Web as Corpus”, Computational Linguistics, 29, 333-47.

Langacker, R.W. (1987). Foundations of Cognitive Grammar. Vol. 1: Theoretical Prerequisites. Stanford: Stanford University Press.

Langacker, R. W. (1991). Foundations of Cognitive Grammar. Vol. 2: Descriptive Application. Stanford: Stanford University Press.

Langacker, R.W. (2008). Cognitive Grammar. A Basic Introduction. Oxford: Oxford University Press.

Rosch, E. (1975). ''Cognitive Representations of Semantic Categories'', Journal of Experimental Psychology: General, 104(3), 192–233.

Taylor, J.R. (1996). Possessives in English. Oxford: Oxford University Press.

Taylor, J.R. (2002). Cognitive Grammar. Oxford: Oxford University Press.

Taylor, J.R. (2003 [1989]). Linguistic Categorization. 3rd ed. Oxford: Oxford University Press.
Ross Deans McLachlan is a PhD student in English Language at the University of Glasgow. His research is primarily in cognitive linguistics and narrative discourse, with an interest in related fields such as cognitive poetics and narratology. His current research project is into the function of first-person plural narration in autobiographies of illness. Focusing on the stylistic effects and rhetorical functions of 'we' narration, this research aims to contribute from a cognitive linguistic perspective to ongoing research into intersubjectivity. His research is funded by the Arts and Humanities Research Council (UK).

Format: Paperback
ISBN-13: 9780199290819
Pages: 336
Prices: U.K. £ 24.99