LINGUIST List 27.1800

Mon Apr 18 2016

Review: Cog Sci; Lang Acq; Neuroling; Psycholing: Li (2015)

Editor for this issue: Sara Couture <>

Date: 26-Feb-2016
From: Geoffrey Sampson <>
Subject: The Handbook of East Asian Psycholinguistics 3 Volume Paperback Set
E-mail this message to a friend

Discuss this message

Book announced at

TITLE: The Handbook of East Asian Psycholinguistics 3 Volume Paperback Set
PUBLISHER: Cambridge University Press
YEAR: 2015

REVIEWER: Geoffrey Sampson, University of Sussex

Reviews Editor: Helen Aristar-Dry


This three-volume work comprises collections of separately-authored chapters on the psycholinguistics of the three leading languages of East Asia: Volumes 1, 2, and 3 cover Chinese, Japanese, and Korean respectively. The first two were originally published, in hardback, in 2006, and the Korean volume in 2009; they went into paperback as a set in 2015. The editors suggest that these three languages may be the only non-Indo-European languages to have attracted significant psycholinguistic research interest to date.

Most of the authors (except in the Chinese case, the overwhelming majority) are native speakers of the languages they write about, though there is also a sprinkling of Western authors and co-authors, and quite a few of the Orientals are based in Western institutions.

The Korean volume is not only newer but much longer than the other two. This is perhaps surprising, considering that Korean is easily the smallest of the three languages in speaker numbers; the volume has more chapters than the Chinese volume, and the chapters are longer than in the Japanese volume. (Precise figures, excluding prelims, are: Chinese, 32 chapters, 455 pp.; Japanese, 44 chapters, 409 pp.; Korean, 44 chapters, 638 pp.)

In order to give the volumes some internal structure, the chapters are grouped into sections on “Language acquisition”, “Language processing”, and, in the Chinese case, “Language and the brain” – though these divisions are not particularly rigid, for instance the first chapter in the Chinese “Language and the brain” section, by Terry Kit-fong Au, is really about the so-called Sapir–Whorf hypothesis. Au’s chapter is very interesting (he argues that Alfred Bloom’s famous 1981 claim that Chinese lacks grammatical support for hypothetical thinking rested on an inadequate knowledge of spoken Chinese idiom); but it is not about brain mechanisms, as some of the following chapters are. (For instance, Li Hai Tan & Wai Ting Siok find measurable differences in language-related areas of brain anatomy between native speakers of English and Chinese, which they attribute to different processes involved in acquiring the respective languages.)

Within the bounds of this review I cannot undertake to list every individual subject covered in the 120 chapters of the three volumes, but I hope that the topics I do take up will give readers a reasonable impression of the books’ overall coverage, which is very extensive.

All three languages discussed contrast with European languages in many ways. Chinese is the language of the world’s oldest surviving civilization, and the language with by far the most speakers in the world. It is much closer than any European language to the pure “isolating” type, with very little morphophonemic alternation. It is genetically unrelated to the languages of the other two volumes; but, because most aspects of the relatively young cultures of Korea and Japan derived ultimately from China, Chinese has had a massive impact on those languages. As a source of vocabulary, Chinese is more significant for both Korean and Japanese than Latin and Greek together are for English (and it has influenced them in other ways too). Korean and Japanese are believed probably to be related to each other, though if so the relationship is a distant one. Both are agglutinating languages; an important grammatical category for both Korean and Japanese is relative social status of speaker and hearer. Korean has traces of having been a vowel-harmony language in the not very distant past, and it has complex morphophonemic rules which create differences between surface and underlying phonological forms that seem large by comparison with any European language. Korean and Japanese are verb-final; Chinese is sometimes seen as in transition from VO to verb-final. All three languages have an astonishingly high incidence of homophones; in Chinese there are very few morphemes which are not homophonous with at least several, often many, other morphemes.

In terms of script, Chinese uses a logographic script in which, essentially, morphemes are in a one-to-one relationship with written forms. Korean is nowadays written mainly in a phonographic script in which, unlike in European alphabetic writing, the separate phonetic features comprised in a consonant or vowel segment are independently represented in the written form of that segment, and those written forms are grouped into syllable (rather than word) units. Until recently the many Chinese loans in Korean were written as Chinese graphs; the modern trend is to use the phonographic script for them too, but this creates problems for readers because of the high incidence of homophones referred to above. Japanese is written in a remarkably complicated script in which lexical roots are written with Chinese graphs but where, unlike in Chinese, the same graph can represent diverse morphemes in different contexts, and elements of the language not written in Chinese are spelled out in one or other of two syllabic systems.

All in all, it seems clear that if languages differ in the ways in which they are processed psychologically, East Asian languages should offer plenty of scope for describing differences from the heavily-studied languages of Western Europe. And these books certainly do that.

Sometimes, the material here is novel because it relates to structural aspects of East Asian languages for which European languages have no parallel. An example would be chapters on the psycholinguistics of individuator words, commonly called “classifiers” or “measure words”. One way to explain what these are is to say that, in East Asian languages, all nouns are mass nouns. As in English we say “two ounces of gold” or “two ingots of gold” but not, usually, “two golds”, so in Chinese one does not say “two dogs” but rather what might be literally translated as “two _tiao_ of dog”, where _tiao_ is an individuator for dogs, roads, and some other long and wriggly things. A language will have one or a few default individuators that can be used very generally and are acquired at an early age, and perhaps a few dozen more specific individuators, like _tiao_, which are acquired later. It is interesting to learn from Mary S. Erbaugh’s chapter that, in Chinese, one of the specific individuators acquired earliest and most reliably is the one for books, which are surely not the most salient objects in the world of an average Western child. Hyeongjin Lee investigates experimentally whether the grammar of individuators correlates with distinctive mental ontology on the part of Korean-speakers.

Again, European-language linguists are used to the idea that grammar will often be influenced by questions of which pieces of knowledge are or are not shared by speaker and hearer. Haruko Minegishi Cook suggests that one important area of Japanese grammar, essential for competent conversational participation, depends on whether or not emotional attitudes are shared by the participants – something for which I can think of no parallel in European languages. Cook uses children’s early acquisition of this grammar to argue against Piaget’s claim that children universally begin with egocentric speech. And this perhaps links to claims by Toshiki Murase & Tamiko Ogura and by Patricia Clancy that the nature of Japanese and Korean “motherese” is different from Western motherese: “Japanese mothers are more affective or empathy oriented and less information oriented [in speaking to children] than North American mothers”; “Korean mothers use more verbs in active play than American mothers, who rely more on labeling”.

In other cases, the linguistic categories discussed are familiar but these languages offer counter-evidence to what Western linguists have taken as reliable generalizations about language acquisition and behaviour. The point just mentioned about nouns and verbs in motherese connects to one of those. A number of Western linguists have argued that nouns have priority over verbs in language acquisition, with nouns learned earlier and in greater numbers, and they have given reasons why that should be so, sometimes relating the issue to generative ideas about innate linguistic knowledge. However, many of these authors claim that, in East Asian languages, it just is not so, and indeed that verbs take priority over nouns. Yuriko Oshima-Takane uses data on this to argue that children’s language-acquisition is determined by superficial properties of the child’s experience more than by deeper factors relating to the structure of the language acquired. (The facts are complicated, though; You-kyung Chang-Song & Soyeong Pae believe that the Korean data can be reconciled with the Western noun-first generalization.)

Even when the conclusion of a study is that, with respect to the topic examined, the Oriental language(s) in question does or do not differ from European languages, this is often well worth establishing. Leading Western researchers have claimed that eye-fixation times in the reading process are longer for Chinese than for alphabetic scripts, which might not seem surprising, but Gary Feng finds that the durations are not significantly different. People sometimes think of an isolating language like Chinese as a language “without grammar”, and since “specific language impairment” in the West tends to manifest itself particularly in relation to morphology, one might expect that Chinese children would be relatively immune; however, Paul Fletcher, Stephanie Stokes, & Anita M.-Y. Wong find that “learning an isolating language is no antidote to potential grammatical problems”.

The three volumes have separate editors, who seem to have made independent decisions about specific topics to be covered – the volumes are not parallel in that respect. Sometimes this may simply reflect different research priorities among students of the respective languages. For instance, the Japanese volume has a chapter by Yasushi Terao on speech errors, but there are no comparable chapters in the other volumes, and it may be that few Chinese or Korean researchers have chosen to study them. The Korean volume has three chapters, and the Chinese volume one chapter, on aphasia, whereas I noticed no mention of that topic in the Japanese volume. On the other hand, only different choices by editors could explain why there is nothing in the Chinese volume similar to the valuable chapter by Masayuki Asahara, Yasuharu Den, & Yuji Matsumoto on electronic research resources available for Japanese, since it would certainly be possible to compile a similar survey for Chinese. (I do not know the situation in Korea.)


One danger with a book of this kind is that contributors may see it as a chance to put their own pet research projects in the shop window, whereas readers will be hoping for a broader survey of the state of play in the contributor’s field. A few chapters in these volumes do feature an unattractively high proportion of self-citations, but on the whole the authors have done a very good job of providing a conspectus of current knowledge and thinking, making the books an excellent resource for anyone planning to embark on new work in the area, or simply to ascertain the current consensus.

A related problem is that academics deeply versed in a particular research issue may fail to appreciate how much needs to be explained to outsiders. A perfect balance here is unachievable, because there are so many different levels of knowledge that diverse readers may bring to a book. Nevertheless, on the whole I thought these contributors got it about right. My impression was that cases where too much was assumed occurred particularly in the Japanese volume. For instance, Keiko Koda on “Development of lexical competence among second-language readers” discusses differences between English-speaking and Japanese- or Chinese-speaking readers in terms of a concept of “phonological inaccessibility” which is not explained and, to me, not self-explanatory. And Yukie Horiba on “Reading in Japanese as a second language” discusses a writing style apparently common to all three Oriental languages which he calls the _ki-sho-ten-ketsu_ style, without explanation; the romanization is not accompanied by _kanji_, so I have no basis for guessing what this might refer to. However, contributors to the Japanese volume might reasonably excuse their terseness by pointing to the fact that the space allotted to their individual chapters is much less than for the other languages.

The standard of English is in the main excellent, considering that most authors and editors are not native speakers. (I did find some scattered passages particularly in the Korean volume to be difficult or impossible to follow.)

Although these books are heavily concerned with empirical data, inevitably a number of contributions are influenced by generative theories of “Universal Grammar” (UG). When this leads to writing which uses observable facts to argue for or against nativist models, this is very welcome. For instance, Heejeong Ko, Tania Ionin, & Kenneth Wexler use data on the mastering of article usage in English by adult native speakers of Korean (which lacks articles) in order to argue that the L2-acquisition process depends on access to UG, while conversely Soo-Ok Kweon uses such speakers’ acquisition of the English contraction _wanna_ to argue that adult L2 learners have no such access. This sort of controversy is how knowledge advances, I believe.

However there are other contributions, particularly in the Japanese volume, which merely treat UG as axiomatic in a way that seems rather naive. Consider, for instance, the chapter by Edson T. Miyamoto. In Japanese the various case roles of nominal constituents in a clause are explicitly marked by postpositions (there are no roles marked only by position, like subject and direct object in English); and the ordering of the different roles is rather free. However, Miyamoto has found a generative article which claims that, underlying an example of free word order, there will be a grammar which produces constituents in a fixed canonical order, after which an optional rule scrambles them into other orders. Consequently Miyamoto carries out a series of reaction-time experiments designed to show that utterances where scrambling has applied are harder to process than ones where it has not. It is not clear to me that there is any need for a hypothesis as far-reaching as Universal Grammar to explain the timing differences Miyamoto observes, and I see no real reason to think that when case roles occur in a given order in an utterance, they did not occur in that order at all stages of production.

Perhaps it is not yet well known in Japan how sceptical many linguists elsewhere have become about the generative style of linguistics. Scholars of the two other languages clearly do appreciate this. The editors of the Chinese volume begin their introductory chapter by discussing the present-day tension between believers in and sceptics about language universals. The editors of the Korean volume make a similar point in their preface, writing for instance that “The search for universals in language processing is a worthy pursuit, but such universals should be discovered and confirmed, not assumed”; and the lead editor, Chungmin Lee, uses his own chapter on “The acquisition of modality” to argue that the ages at which English- and Korean-speaking children master the grammar of modality are so different as to refute the notion of an innate Language Acquisition Device.

Despite their complexity, I spotted few errors in these volumes. Inevitably there are some. The chapter by Yu-Chin Chien & Barbara Lust contains repeated mistakes in the _pinyin_ romanization of Chinese examples (for instance, ignoring tones, the standard Chinese for ‘tell’ is spelled _gaosu_, not _gaoshu_ or _gaushu_). A remark by Jun Yamada in connexion with statistics on the incidence of dyslexia in different cultures implies that Yamada regards Japanese script as one of the world’s simpler writing systems – pretty well everyone else who has discussed the matter concurs with Richard Sproat’s judgement (2000: 132) that Japanese script is “surely the most complex modern writing system”. Douglas Honorof & Laurie Feldman’s statement that Chinese graphs represent not morphemes but syllables is possibly ambiguous, but as it would be understood by most linguists it is seriously misleading: in a syllabic script, a given phonological syllable is written in a given way irrespective of what word it occurs in, but in Chinese script distinct morphemes are normally written differently even if they are homophones, though with marginal exceptions each morpheme is one syllable long. Examples in a chapter by Chungmin Lee & Sook Whan Cho are glossed and translated inconsistently: a Korean word _tanpwung_ is glossed ‘colored.leaves’ in one example but as just ‘leaves’ in another, and translated in both as ‘fallen leaves’. It was startling to find Nobuhiro Furuyama turning the famous German psychologist Wilhelm Wundt into a Dutchman, “Willem Wunt”. And there are sporadic cases of individual Chinese graphs transliterated with wrong tones, and so forth. But it might be difficult to execute a publishing enterprise on this scale with a much lower incidence of errors.

Turning from the contents of these books to editorial issues: now that it is becoming commoner than it once was for East Asian scholars to publish in English, it strikes me that English-language editors and publishers need to think more than they have in the past about adapting copy-editing standards to cater for East Asian realities.

Consider the issue of personal names. Throughout East Asia, inherited family names precede individuals’ given names, the opposite of practice in most Western countries, though when publishing in English some (but not all) East Asian scholars defer to Western norms by reversing their names, which looks strange to those used to Oriental names. (I have never seen the late Chinese leader referred to as Tsê-tung Mao, or Zedong Mao in modern _pinyin_ transliteration.) Some academics’ names appear in opposite orders in different Western publications. In China there is only a small set of family names (the Chinese cliché is “100 names”, though I believe the true figure is somewhat larger), and they are almost all single syllables, though a few have two syllables. Given names are coined very freely; most often they comprise two syllables, but single-syllable given names are quite common. (Korean personal names, like other aspects of Korean culture, are modelled on Chinese practice. Japanese family and given names are commonly polysyllabic.)

Take the name of the general editor of these volumes, shown on the title page as “Ping Li”. Evidently both his names are monosyllables (I know it is “his”, because the brief bio in the prelims uses the masculine pronoun – from the name alone one could not tell). If the name were shown in Chinese alongside the romanized form, it would be easy to see which of the two names was a family name; but these volumes make only sparing use of Oriental scripts (except in chapters whose topic is reading and learning to read), and we are not shown the Chinese form of “Ping Li”. As it happens, both “Ping” and “Li” are possible romanizations of Chinese family names. My guess is that Li is the family name here (i.e. this is a case where a name has been displayed in Western sequence), simply because Li is a much commoner family name than Ping. If that is right, then the general editor would appear in the index or the list of references among the Ls, as “Li, Ping” – with a comma that looks spurious to an Orientalist, since “Li Ping” would be the standard form of the name. But of course my guess could be wrong, and the place to look for the general editor might be among the Ps.

It gets worse when we come to literature citations. For instance, p. 6 of vol. 1 cites “Chen and Peng (1994)”, giving only the authors’ family names, as is usual in Western scholarly publishing. However, Chen is another very common Chinese family name, and the list of references at the end of the volume has almost two pages of Chen entries. The only way to locate Chen and Peng (1994) is to wade through these pages entry by entry, starting with “Chen, Baocun, Chen, Guicheng, Chen, Hao, & Zhang, Zaizhan (eds). (1988)” – note that these are four people, not eight – and continuing with “Chen, E.-S. (2002)” and “Chen, E.-S. (2003)”, until eventually “Chen, Y., & Peng, D.-L. (1994)” turns out to be the very last Chen entry. This is wretchedly time-consuming.

The muddles that arise are many. I tried to check the academic credentials of the contributor Li Hai Tan, mentioned above, but in the “Notes on contributors” he (she?) is listed neither under the Ls nor under the Ts (nor the Hs) – an omission very likely to be connected to the unclarity about where he or she _should_ be listed.

Obviously, Western reference practices have been influenced by the fact that a Western person’s family name is a highly distinctive part of his whole name. This does not work for East Asia. (It would work even less well if this set of books covered Vietnamese; more than half of all Vietnamese share the same family name.) Surely it is desirable to adapt Western conventions to make readers’ lives easier? In my own writing I have found that the only practical approach is to list East Asian names always in East Asian family-name-first order (without spurious commas), irrespective of how the person has chosen to present his name in a particular English-language publication, and to include given names or at least initials following East Asian family names in literature citations. (In this review, though, I have shown names as they appear in the volumes reviewed; one cannot standardize on family-name-first when it is not always clear which is the family name.) To date, Oriental academics seem to have seen it as more important to conform rigidly to every detail of the conventions evolved within Western scholarly practice than to adapt those conventions to their own realities. As East Asians become more frequent contributors to the international republic of scholarship, it is to be hoped that they will be more inclined to modify unsuitable Western conventions.

But let me not end my review on a carping note. This set of books is an admirable achievement. It will surely become a landmark in its field for many years to come.


Bloom, A.H. 1981. The Linguistic Shaping of Thought: a study in the impact of language on thinking in China and the West. Lawrence Erlbaum (Hillsdale, N.J.).

Sproat, R. 2000. A Computational Theory of Writing Systems. Cambridge University Press.


Geoffrey Sampson graduated in Oriental Studies at Cambridge University in 1965, and studied Linguistics and Computer Science as a graduate student at Yale University before teaching at the universities of Oxford, LSE, Lancaster, Leeds, and Sussex. After retiring from his Computing chair at Sussex he spent several years as a research fellow in Linguistics at the University of South Africa. Sampson has published in most areas of Linguistics and on a number of other subjects. His most recent book is a new edition of ''Writing Systems'' (Equinox, 2015).

Page Updated: 18-Apr-2016