Review of The Handbook of Linguistics, second edn. Edited by Mark Aronoff and Janie Rees-Miller. Wiley Blackwell, 2017.

Reviewed by Geoffrey Sampson

To appear on the LINGUIST List, <www.linguistlist.org>

SUMMARY

The editorial introduction to the first edition of this book, which appeared in 2001, described it as not a student textbook but rather “an authoritative volume on what linguists know about language at the start of the twenty-first century”, intended for “people who would like to know what linguistics and its subdisciplines are about. The book was designed to be as nontechnical as possible”. The new edition announces that its goals remain much the same, but the book has been updated to reflect developments over the past twenty years. I have not seen the original edition so cannot compare them directly (its introduction, quoted above, is reprinted in the new edition), but the editors tell us that all but five of the 33 chapters are either revised or entirely new, often by new contributors.

A list of 33 chapter titles would make for an unreadable review, though the majority of individual chapters and authors will be mentioned in the course of what follows. However, the chapters are grouped into the following sections:

Part I Starting points (5 chapters)

Part II Theoretical bases (3 chapters)

Part III Core fields (7 chapters)

Part IV Languages and the mind (4 chapters)

Part V Language in use (8 chapters)

Part VI Applications of linguistics (6 chapters)

In the case of Part VI it is worth noting that whereas in Britain “applied linguistics” is normally shorthand for “linguistics in the service of teaching English as a foreign language”, here several other applications are also covered: for instance there is a chapter by Roger Shuy on language and law, and one by Kathryn Stemper and Kendall King on language planning. Most chapters are about ten thousand words long, though some in Part VI are shorter (explained by the editors as due to these topics being more specialized).

Taken as a whole the book covers the subfields of linguistics rather comprehensively. One notable gap is that there is no chapter on computational techniques in linguistics, and only brief passages within a few of the chapters on applications of computing to their topics.

The book is strongly oriented to linguistics as practised in the USA. Both editors are based in America, and so are two-thirds of the 36 contributors. There are also seven contributors from Britain, and one each from Canada, Germany, Sweden, Israel, and New Zealand.

Apart from lists of works cited at the end of each chapter, some chapters have lists of “suggestions for further reading”, “relevant journals and societies”, and/or “emergent trends and research questions”. I find no particular pattern in which chapters include some of these lists and which lack them; perhaps their inclusion was a suggestion by the editors which some contributors found it convenient to comply with and others did not.

With a book in which many different authors cover such a wide range of topics, once one has given these largely statistical facts it is difficult to say much more as a précis of the book contents. In this review, therefore, I keep the “Summary” section short and describe the nature of the contents more fully in the course of evaluating the book in the next section.

EVALUATION

A first criterion for a book of this kind must be whether the facts it presents are true facts. In this respect the book is not flawless.

J.K. Chambers (“Sociolinguistic theory”) is clearly wrong to attribute an (ungrammatical) quotation to “p. 365” of a 250-page book by the philosopher William James, and I believe Chambers may actually be quoting a loose paraphrase of James’s ideas by the author of the novel “Zen and the Art of Motorcycle Maintenance”. Chambers also overgeneralizes wildly in claiming that, in societies with a marked division of labour between the sexes, it is “invariably” women who negotiate bank loans. England in the nineteenth century was such a society, and before 1882 a married Englishwoman could not own property or enter into a contract; so how could she have applied for a bank loan? Contrary to Peter Daniels (“Writing systems”), the Modern Greek /b/ sound is written μπ, not μβ (which represents /mv/). The first example in Brian MacWhinney’s “First language acquisition” only works if ‘who’ is inserted after ‘boy’ in his sentence (1c). Contrary to Shlomo Lappin (“Formal semantics”), in terms of philosophical logic the extension of a relation is not a sequence of objects but a set of sequences; and an assertion about a non-existent object (instead of the standard example, ‘the present King of France’, Lappin uses ‘the 42nd US President’ uttered during the administration of the 41st president) is usually seen not as false but as lacking truth-value. In Vivian Cook’s “Second language acquisition”, in order to exemplify the point he is aiming to make (and in order to be good German), the example ‘Da Kinder spielen’ ought to read ‘Da spielen Kinder’; and the Sino-Korean word for “colour” is ‘saek’, not ‘sekj’. Suzanne Romaine (“Multilingualism”) claims that only six EU member states, which she lists, are officially bi- or multi-lingual, but Stemper and King point out that Sweden (not on that list) made Saami (Lappish) an official language twenty years ago. Furthermore, when the book was published, Britain was a member state. English law does not have a concept of “official language”, but a British passport is written in Welsh and Scots Gaelic as well as English; any public notice or communication in Wales has been bilingual for decades, and the same is increasingly true now in Scotland and Northern Ireland – there have even been attempts to elevate “Ulster Scots” into the role of public written language.

At a deeper level, Brian Joseph’s “Historical linguistics” chapter is surely mistaken to say that, because the history of a language can be conceptualized as a series of synchronic states, the problem of determining what range of language changes is possible reduces to that of determining what is synchronically a possible human language. English has words beginning with /b-/ and words beginning with /g-/, and I suppose a language like English except that the /b-/ words all began instead with /g-/, and vice versa, would be an equally possible language; but it at least does not logically follow that a sound-change which swapped initial /b/’s and /g/’s throughout would be a humanly possible change (and my guess is that it would not be possible).

Still, this is arguably not a high incidence of factual errors in a book of almost 700 pages. And some of the chapters offer very useful surveys of what is currently known about their special field. Two chapters I found specially valuable, for instance, were David Caplan on “Neurolinguistics” and Kiel Christianson on “Psycholinguistics” – because these subjects advance through numerous experiments reported in separate journal articles, it is difficult for an outsider to gain an overall impression of the current state of knowledge, and these chapters fill that gap very well.

Nevertheless, on balance the book disappointed me.

Since its declared purpose is to portray linguistics to intelligent outsiders wanting to know what the subject has achieved, I tried to read it as if I were in that situation, perhaps as a young person choosing a subject to study at university.

The impressions formed by such a reader would be heavily coloured by the contents of the earlier chapters, including the “Core fields” section (indeed, one wonders how many readers of that type would manage to get all the way through this long book). The reader would find those chapters very abstract, and might wonder about standards of evidence within the discipline. To take one example, for Mark Baker’s “Syntax” chapter it is important to establish that a transitive clause has a bipartite subject–predicate structure, with object linked to verb more closely than either is to subject. To demonstrate that, he claims that the clause ‘John likes that dress’ can be altered to ‘Like that dress, John does’, but not to ‘John like, does that dress’. Perhaps not; but one can say ‘That dress, John likes’ – indeed this sounds more natural to me than Baker’s version. And in another respect, number and person agreement, it is the subject rather than object which has a special relationship with the verb. Baker gives no hint why those facts should be less relevant to grammatical structure than the facts he cites. Yet these are points likely to occur to any lively-minded undergraduate.

Lyle Campbell (“The history of linguistics”) works through the history of how Chomsky’s generative grammar replaced simpler syntactic models with models involving grammatical transformations, and then with further models. Mark Baker alludes briefly at one point to the fact that Gerald Gazdar’s Generalized Phrase Structure Grammar eliminated the need for much of this complex grammatical machinery, but that is not followed up anywhere in the book, and Baker rehearses the history of Chomsky’s changing models again – actually for the third time, because it has also been done by Thomas Wasow (“Generative grammar”). Books introducing a subject to newcomers do not usually feel bound to spend time on discarded ideas; the reader may wonder whether linguistics is the study of language, or of one man’s intellectual biography. Andrew Carstairs-McCarthy (“Origins of language”) tells us that “Chomsky for a long time discouraged interest in language evolution”, but “In 2002, this situation changed dramatically with the publication of an article … by Chomsky and the animal behavior experts Marc Hauser and Tecumseh Fitch … Since then, linguists associated with Chomsky have been willing to discuss language evolution”.

There are many allusions to “language universals” and “Universal Grammar”, but without much explanation of what these terms refer to, let alone what reason one might have for believing in them. The earliest reference I found to UG, for instance, was in William Croft’s “Typology and universals” chapter, where he says “The existence of … implicational universals requires a rethinking of the nature of Universal Grammar”. The intended readership is unlikely to be aware at this point of UG as a concept, still less to associate enough content with the term to be able to “rethink” it. And that is not just because the intended readers have no prior linguistics background; Steven Pinker, himself very much in sympathy with the style of linguistics represented here, remarked that “UG has been poorly documented and defended in the linguistics literature” (Pinker 1998), and I do not know that that situation has changed much in the subsequent twenty years.

Thomas Wasow cites the “poverty of the stimulus” as evidence for innate language universals – that is, the alleged fact that the experience available to children acquiring their mother tongue typically contains no evidence about features which the children nevertheless succeed in mastering. But Wasow gives no example of such a feature. This did not surprise me, because I do not believe there are any examples. The English-language case usually cited in the literature is the supposed absence from the child’s experience of a particular type of question; but these questions turn out, empirically, to be fairly common in the kind of casual speech to which children are exposed (Sampson 2005: 79ff.).

(The clearest attempt I found in the book to make Universal Grammar stand up was not in the early theoretical chapters but in Wendy Sandler and Diane Lillo-Martin’s much later “Sign languages” chapter, which claims that American Sign Language obeys Haj Ross’s Co-ordinate Structure Constraint. Viewed through the lens of Generalized Phrase Structure Grammar, though, Ross’s constraint does not offer much support for the idea of innate knowledge of language.)

The theoretical abstractions which are central to much of the book are presented as facts agreed by general consensus, with very little suggestion that they are often highly controversial. Ironically, authors quoted as significant contributors to the development of these theories include several references to Paul Postal, Geoffrey Pullum, and Philip Lieberman – all of whom in their younger days did make such contributions, but all of whom later became opponents, often outspoken opponents, of the style of linguistics represented here – and in Lieberman’s case, of the very idea which Carstairs-McCarthy describes him as supporting. Only the early publications of these writers are cited, never the dissenting ones. (See e.g. Postal 2004, Kornai and Pullum 1990, Pullum 1996, Lieberman 2000.)

Several contributors discuss the “Proto-World” idea, that all human languages share a single common ancestor, as if this were an agreed consensus; that seems a seriously misleading treatment of a topic likely to interest the general public. I believe the consensus among knowledgeable historical linguists is much closer to the position of my late colleague Larry Trask, that there probably was not one common ancestor, and if there was, evidence for it must long ago have been irretrievably lost. (Cf. Campbell 2008.) Carstairs-McCarthy even argues here that all languages must share a common ancestor, because otherwise we would find individuals of different races unable to learn one another’s languages – an argument which puzzled me greatly until I realized that Carstairs-McCarthy was tacitly assuming that the birth of language was triggered by a genetic mutation for a specific type of language.

One indication of the partisan nature of much of the book is that Michael Tomasello’s name does not appear in the index. I did find one brief mention of Tomasello: Vivian Cook cites him in passing as the kind of researcher “that linguists have loved to scorn”. Some linguists, perhaps; but it might occur to a reader that scorn can be a two-way street.

Readers of earlier chapters of the book will find linguistics to be heavily concerned with abstruse terminology and symbols, despite the editors’ declared intention to make the book nontechnical. Abigail Cohn (“Phonology”), for instance, describes hypothetical constraints on how different from one another underlying and surface phonological shapes are allowed to be; one constraint says “don’t add material to the input” and another says “don’t delete material from the input”. These phrases are clear enough in themselves, and for newcomers the author’s decision to represent them with symbols, *ADD and *DELETE, might be more confusing than helpful; not only does she do that, but she then adds that they are standardly called “DepIO” and “MaxIO”. Steven Black and Elizabeth Falconi (“Linguistic anthropology and ethnolinguistics”), after beginning their chapter by saying that their ration of space is too short to cover their topic adequately, then spend their next section discussing how that topic has been given different names within different academic traditions. Readers might think it a higher priority to learn what concrete findings have been discovered, under any name.

One key to the unsatisfactory nature of this book is a passage in the editorial introduction, which defines linguistics as “the scientific study of language”, notes that people often see language and other human activities as “beyond the scope of true science”, but responds: “it should be possible to have a science of anything”. If “science” is taken in its usual English-language sense, this proposition is not just profoundly mistaken, but a surprising one to be espoused by 21st-century thinkers. Leading philosophers of science do not agree. Sir Karl Popper’s famous “demarcation criterion” was not a boundary between science and nonsense, but between science and other subjects, some of which are valid and very worthwhile, but not accessible to scientific method. In the early twentieth century there was an influential philosophical trend, logical positivism, which did hold that no discourse was meaningful unless it could be reduced to propositions of empirical science, but few have much time for logical positivism today; John Passmore (1967: 56) described it as “dead, or as dead as a philosophical movement ever becomes”. Yet within linguistics scientism lives on, though not often made as explicit as in these editors’ remark, and to my mind it is responsible for many of the shortcomings of linguistics as it has developed in the 21st century (Sampson 2017). Some important aspects of language are indeed “beyond the scope of true science”, so if they are forced into a scientistic mould perhaps it is inevitable that the result is a pastiche of unnecessary formalisms and unfounded theorizing.

Some chapters of the book, particularly in later sections but also e.g. Bernard Comrie on “Languages of the world”, Pamela Munro on “Field linguistics”, or John Laver on “Linguistic phonetics”, are solid contributions which escape the criticisms made here. But these chapters do not set the tone of the whole. (Pamela Munro even includes a woebegone suggestion that “theoretically oriented linguists hold those who collect primary data in low esteem”.) Young university entrants, or civil servants deciding which academic areas deserve public funding, would learn from this book that linguistics scholars are given to ex cathedra theoretical pronouncements, hung up on discipline-internal formalisms and terminology, unwilling to think new thoughts until given the go-ahead by their master, and out of touch with intellectual currents in the wider world. They might decide to give this discipline a miss. I wish I felt more confident that these aspects of the book misrepresent 21st-century linguistics.

REFERENCES

Campbell, L. 2008. “What can we learn about the earliest human language by comparing languages known today?” In B. Laks, ed., Origins and Evolution of Language, Equinox (Sheffield), 79–111.

Kornai, A. and G.K. Pullum. 1990. “The X-bar theory of phrase structure”. Language 66.24–50.

Lieberman, P. 2000. Human Language and Our Reptilian Brain. Harvard University Press.

Passmore, J. 1967. “Logical positivism”. In P. Edwards, ed., The Encyclopedia of Philosophy, vol. 5, Collier Macmillan (London), 52–7.

Pinker, S. 1998. Posting 9.1209 on the LINGUIST List, <www.linguistlist.org>, 1 Sep 1998.

Postal, P.M. 2004. Skeptical Linguistic Essays. Oxford University Press.

Pullum, G.K. 1996. “Nostalgic views from Building 20”. Journal of Linguistics 32.137–47.

Sampson, G.R. 2005. The “Language Instinct” Debate, revised edn. Continuum (London and New York).

Sampson, G.R. 2017. The Linguistics Delusion. Equinox (Sheffield).