The two books reviewed here provide two counterpoints to each other. Geoffrey Sampson’s Linguistics Delusion argues that linguistics is not and cannot be a scientific study of language, because language is too creative to be systematically studied and theorized about in the same way that hard sciences like physics and chemistry are able to theorize about their subject matter. This book is a weighty read and likely not approachable for a casual reader or a novice in the field. On the other hand, Geoffrey K. Pullum’s Linguistics: Why it Matters provides a case for the importance of the study of language for readers outside the field and offers an approachable and reader-friendly text. The target audiences of the two books are quite different, but the spirit of the books – the point of linguistics – is similar.
The Linguistics Delusion is divided into four sections and thirteen chapters, which had originated from essays that Sampson had previously written and now revised for the current work. In each section, he tackles a different assumption of disciplines of linguistics. In Chapter 1, Sampson delivers his argument: ‘Human language is not the kind of thing that can be studied using the methods of science’ (3), and ‘Linguistics is not a science’ (5). Instead, linguistics should, at most, be firmly ensconced within the humanities or arts disciplines, and perhaps should not ‘rank as a “subject” at all’ (5) as a coherent body of knowledge taught in universities. He acknowledges that linguists will likely find these claims extremely contentious, and so will spend the remainder of the book defending them.
In the first section, ‘Language over-theorized’, he sets up his thesis and discusses several linguistic theories within the scope of Chomskyan generative linguistics and their pitfalls. Chomsky’s Syntactic Structures and Topics in the Theory of Generative Grammar are frequently referenced as Sampson builds his arguments against grammaticality judgments and generative syntax (Chapters 3 and 4); Gricean maxims are referenced as inadequate for explaining the nature of conversations that take place (Chapter 7). He also suggests that linguists have something to learn from economic theory with respect to theorizing human behavior (Chapter 5), and also discusses the flaws of embodied cognition and other cognitive theories of language.
Chapter 2 begins with the main argument against linguistics as a field: namely, that humans’ language use is too creative to be defined in a set of scientific laws. The theories that have been posited so far are primarily based on English and other Eurocentric languages, so provide inadequate explanations for other languages’ structure; and the types of structures that we see used do not conform to computational models of language use. To better explain the differences between the type of creativity that can be scientifically explored as opposed to activities that are ‘creative in the traditional sense’ (26), such as that found in paintings and novels, Sampson uses the terms F-creative (novel and creative scientific theories) and E-creative (artistic works). Exploits that are F-creative should generate a specific set of phenomena that can be falsified, while those that are E-creative may be able to be investigated systematically, but the set of predictions generated by the systematic investigation cannot be falsified.
Chapters 3 and 4 discuss the problematic nature of grammaticality judgments and innate grammar, respectively. Anyone who has taught syntax and used sample sentences to demonstrate aspects of language that are not grammatical for native speakers has likely also run into the problem of students claiming that the sample sentence is, in fact, grammatical for them. Sampson points out that with these types of ambiguously grammatical sentences, there must be some cutoff point to define what is acceptable and what is not, and that, by its very nature, would have to be an arbitrary distinction. This feeds into the notion that because there is so much variation in what is considered acceptable, it would be very difficult to parameterize an innate grammatical system, à la the principles and parameters framework. Instead, Sampson opts for a more behaviorist approach.
In Chapter 5, he explores the economic theory of endogenous growth and how it may be applied to linguistics. He relates endogenous economic growth to the logical outcome of Chomsky’s principles and parameters, which should suggest that there is a finite and bounded possible set of languages based on having a finite set of parameters. He asks us to consider whether human cognition is truly open-ended, or if many of the great ideas have already been thought of, and the rest of us engaging in scientific enterprises have to sort through the detritus to come up with smaller important ideas.
Chapters 6 and 7 deal with the struggle of theorizing meaning and cognition in a suitably cross-linguistic fashion. Sampson points out that languages have different ways of conceiving of the passage of time, among other things, and critiques the notion of a universal set of concepts to enable theorizing about meaning without referring to language-internal relationships. In Chapter 7 in particular, he criticizes Gricean maxims for being widely thought of as a good overarching theory of conversation when they may not be. The example he uses is an interaction between an interrogator and a spy being able to convey something meaningful back and forth, but without adhering to Gricean maxims.
Sampson addresses writing systems in the second section. Over Chapters 8 and 9, his aim is to discuss why writing systems tend to be neglected by theoretical linguists and why they should not be. Sampson draws on his knowledge of Chinese orthography and writing systems in general to critique some previous research generalizing about writing systems. In both chapters he discusses Chinese orthography in particular detail, providing arguments against phonemic alphabets in Chapter 8, and evidence against compound ideographs in Chapter 9.
The third section, ‘Language complexity’, addresses complexity in language, especially regarding the frequent claim that all languages are equally complex. In Chapter 10, he refers to two studies purporting to analyze linguistic complexity across many languages, and points out that they find very different results for three of the languages that both studies tested. He does not address the fact that both studies used different measures of complexity and different theoretical assumptions, so there may be more at play than this book claims. In Chapter 11, Sampson attempts to argue that language differs in levels of complexity as a cultural by-product by comparing linguistic complexity to the complexity found in different types of law and how they are interpreted, as further instances of culture-derived complexity.
The fourth and final section discusses phonological change (Chapter 12) and the question of whether there is a finite number of possible trade names (Chapter 13). Sampson points out that if researchers are claiming that phonological change moves in such a way as to avoid homophony or general similarity in form and sound, there are multiple counterexamples with a good deal of homophony in Chinese. In Chapter 13, he tackles the question of whether we really are running out of trade names (such as Advil, the trade name for the drug ibuprofen). He does this by computing the number of possible consonantvowel combinations, with a few constraints, and finds that there is in fact a finite set of good-sounding English trade names, which could be names like Rallyhack or Footswoll . The set is very large, but finite nonetheless. The author presents this as an argument in favor of linguistics, or at least an application of linguistics that is useful.
Linguistics: Why it Matters consists of five chapters and a conclusion. It should be noted here that this series of books is intended as brief introductions to various subjects, and the publisher enforces a word limit of 25,000 words. Thus, Pullum is not able to go into a large amount of detail while he describes the state of the field of linguistics and why non-linguists should care about it. That said, this book is a clear and concise overview of what linguists care about and how they study language, as well as the assumptions that they make and why.
Chapter 1, ‘What makes us human’, clarifies the difference between language and communication, details some of the similarities between languages and dialects, mentions the importance of language revitalization, and presents a short case study of one instance of linguistic research. The snapshot of linguistic research focuses on Derbyshire’s study of word order in Hixkaryana, which is illustrative not only of the research process but also explains concepts like word order and the falsification of claims previously made in linguistics. Pullum also notes that linguistics as an academic field is very young (vii) and thorough research on language takes a very long time, so many discoveries about language are still being made. This is contrary to Sampson’s view, which posits that language as a phenomenon has been studied for centuries, by both ancient and modern philosophers.
Chapter 2, ‘How sentences work’, is broadly about syntax, and Pullum takes special care to note that due to prescriptivist English teachers, non-linguists are often convinced that they use bad grammar. He highlights the importance of interpreting sentence structure using an example of a lawsuit resulting from the lack of a serial comma in written legislation (31–35), making the intended meaning of the sentence ambiguous.
Chapter 3, ‘Words, meaning, and thought’, includes a brief section on the Sapir–Whorf hypothesis, lists of so-called untranslatable words that crop up all over the internet, and conversational implicature. Given that this is one of the ways that Sampson attests that language is creative and complex, this chapter can be seen as as a basic rebuttal to Sampson’s chapters on linguistic relativism and his very literal interpretation of Gricean maxims.
Chapter 4, ‘Language and social life’, introduces sociolinguistics and the ways that language use and language register can have an impact on how people are perceived, with particular attention to John Rickford and Sharese King’s research on Rachel Jeantel’s testimony on the murder of Trayvon Martin (see Rickford & King Reference Rickford and King2016). He also includes a section on signed language to illustrate not only the idea that language is a very human phenomenon (in comparison to efforts to teach apes language), but also that varieties of language from African American English to signed languages have predictable rules and patterns.
Finally, Chapter 5, ‘Machines that understand us’, discusses speech recognition and speech technology, and how computers are able to interpret acoustic information to produce a list of places to eat in the general vicinity or what tomorrow’s weather will be like, or how Google understands that you are looking for a particular response when you search for ‘what is the square root of 6389’. This chapter also in part speaks to Sampson’s frustration with trying to find a grammar that computers can effectively use to parse speech; Pullum points out the relative uselessness of Microsoft Word’s automatic grammar checker, which is based on simple algorithms. Both authors agree that the difficulties faced by natural language processing are extremely complex.
In sum, I would be interested in Sampson’s thoughts regarding the extensive body of processing literature that he has largely not addressed in his book, particularly in psycholinguistics, sociolinguistics, and communication disorders. Whether or not linguistics ought to be broken apart into a variety of academic departments instead of retaining its status as a university subject is a discussion I will leave to others, but I am sympathetic to Sampson’s claim that the various theoretical approaches he discusses are inadequate for explaining how language is used in everyday life. This is why one can find linguists in a variety of disciplines – in university linguistics, English, psychology, computer science, communication disorders, and anthropology departments as well as in industry, working for editing firms, advertising firms, small start-ups, and large companies like Google – with a variety of research interests and methodological expertise. Scientific research in linguistics can run the gamut from descriptive analyses to assessing statistical probabilities of studied phenomena, to treatments for language disorders. Pullum describes the interdisciplinary concerns of linguistics as a field coherently and concisely, in ways that are likely not controversial for linguists but might be remarkable to language enthusiasts.
