This book presents a new theory of grammatical categories - the Universal Spine Hypothesis - and reinforces generative notions of Universal Grammar while accommodating insights from linguistic typology.
Date: Sat, 4 Jun 2005 21:35:17 +0100 From: Heather Marsden <H.L.Marsden@newcastle.ac.uk> Subject: Universal Grammar in Second-Language Acquisition: A History
AUTHOR: Thomas, Margaret TITLE: Universal Grammar in Second-Language Acquisition SUBTITLE: A History SERIES: Routledge Studies in the History of Linguistics YEAR: 2004 PUBLISHER: Routledge (Taylor and Francis)
Heather Marsden, School of English Literature, Language and Linguistics, University of Newcastle Upon Tyne
The aim of this book is to explore the relationship through western linguistic history of the two concepts named in the book's title: universal grammar and second language acquisition. Beginning in ancient Greece, the book tracks ideas about commonalities among languages, and about the experience of non-native language learning, through the centuries, culminating with Chomskyan universal grammar and modern second language acquisition theory.
The book's intended primary readership is, as Thomas states (p.18) 'linguists or applied linguists concerned with the theoretical implications of L2 acquisition'. In addition, it is aimed at language teachers and teacher educators, and finally, historians of linguistics.
Chapter 1 introduces the topics and the rationale of the book. The terms 'universal grammar' and 'second language acquisition' are defined here, in the context of the book. The former refers to ideas -- explicitly formulated or not -- about what is universal in human language, as opposed to what is particular to one language or another. The latter is used in a broad sense to refer to acquisition or learning (these terms are used interchangeably) of a non-native language, without differentiating between 'second' and 'foreign' language acquisition. A key goal of the history -- beyond documenting a relationship between ideas that has not been documented before -- is to broaden the horizons of the book's main intended readership: those engaged in contemporary second language acquisition research. Thomas shows how the field of second language acquisition research represents itself as having no history prior to the latter half of the twentieth century: in other words, no history prior to Chomskyan linguistics and the concept of universal grammar in its most modern instantiation. She argues that this ahistoricity creates a false, and potentially detrimental, sense of the nature of second language acquisition research.
The chronological exploration of the history of universal grammar and second language acquisition begins in Chapter 2, which focuses on ancient Greece and Rome. The ancient Greeks produced grammars of Greek, the earliest of which probably dates from the second century BCE. It focuses on describing morphology, parts of speech, and the Greek sound system. The Greeks did not write about second language acquisition; nor did they propose language universals. However, Greek philosophical ideas about Forms (eidos) are argued to initiate the concepts that have later been used to develop ideas about universal grammar. For example, in Aristotelean theory of Forms, the essence of an object is the set of characteristics contained within all exemplars of that object. Applied to language, the essence of grammar would be the linguistic properties shared by all languages: universal grammar.
Roman grammars of Latin followed the Greek model, but they served as textbooks for school pupils studying their L1. Two important grammars continued to be influential into the Middle Ages: Ars Grammatica, written by Aelius Donatus, and Institutiones Grammaticae, written by Priscian of Caesaria. A key linguistic difference between Roman and Greek society was that educated Romans at the height of the Roman empire were bilingual in Greek and Latin from birth, due to the presence of Greek slaves in Roman households. Literacy in the two languages was an essential element of Roman education. However, since children entered school already fluent in both languages, there was no motivation to consider the nature of second language learning. Nor did the Romans investigate language universals, although, unlike the Greeks, they engaged in the comparative study of Latin and Greek.
Chapter 3 focuses on the period from around the fourth century BCE to the end of the first millennium. A key thinker early on in this period was Augustine, whose writings suggest an early conceptualisation of innate universal linguistic knowledge, and also include thoughts on child language acquisition. He contrasts his experience of L1 (Latin) acquisition, acquired simply through observation, with his experience of being forced to learn Greek in the classroom. By implication, L1 and L2 acquisition may have been epistemologically identical to Augustine, but the manner of exposure to the different languages resulted in different outcomes.
Christianity became very influential during this period. Christianity arose in a multilingual context, and, effectively institutionalised L2 learning as a means of furthering its aim of making its message accessible to everyone. By the seventh and eighth century CE, Latin -- as the language of the Western European church -- was being studied as an L2 by Christian speakers of Germanic, Celtic, and Mediterranean languages. Donatus's Ars Grammatica and Priscian's Institutiones Grammaticae were key textbooks, but these were revised and supplemented in order to better meet the needs of L2 learners. In particular, where the Roman authors had assumed L1 knowledge of aspects of Latin, sixth and seventh century authors attempted to spell out the relevant details. Thus, the first true L2 grammars were created.
Chapter 4 covers development of medieval grammatical theory and ideas about L2 learning from the eleventh to the fourteenth century. This period saw the first articulation of the concept of universal grammar. At the beginning of this period, grammar was re-defined as -- in the terminology of the recently rediscovered Aristotle -- a speculative science, meaning a theoretical science which pursues knowledge for its own sake, rather than to a practical end. Groups of speculative grammarians emerged in European university towns in the thirteenth and fourteenth centuries. The assumption that universal grammar exists was core to their work. This assumption derived from their conviction that language is shaped by cognition and the nature of the world. Their aim was to discover as much as possible about the properties of universal grammar through the study of Latin.
L2 acquisition during this period continued to be dominated by the study of Latin, although contact with other languages via travel, commerce, and the crusades was beginning to increase. L2 pedagogy (i.e., Latin teaching) was mostly consistent with the previous period: students had to parse and memorise chunks of L2 text. However, an important development was that the students' native language began to be used extensively in the classroom. Previously, the L1 had been used only at the most elementary level, but, by the late fourteenth century, it was also being used to facilitate advanced L2 learning contexts.
Thomas highlights an intriguing point of interaction between speculative grammar and second language acquisition: namely, that the focus of enquiry into linguistic universals, namely Latin, was an L2 for all scholars. Medieval grammarians did not question whether intuitions about their L2 could lead to insights about universal grammar. They rejected investigation of their native vernaculars since the latter were considered 'intrinsically unruly, structurally defective, and impoverished relative to Latin' (pp. 72-3). This contrasts strikingly with the focus on L1 intuitions, and the scepticism about L2 intuitions, in modern linguistic enquiry.
Chapter 5 covers the period from the beginning of the fourteenth century until the end of the seventeenth. Speculative grammar was gradually abandoned at the beginning of this period. The focus of language study turned back to the description of particular languages -- increasingly vernaculars, and not Latin. The concept of 'living' and 'dead' languages was born (i.e., the vernaculars v. Latin), leading to the emergence of a new framework for comparative language study. Numerous L2 and L1 grammars were published during this period, and the range of languages known to European scholars increased vastly, through contact with the Americas and the Far East. However, the model for analysis of all these languages continued to be that of Donatus and Priscian.
As a reaction against the focus on particular languages, some scholars returned to the ideas of Aristotle, and to universals. Among these was Sanchez de las Brozas (1523-1601), whose logic-based work on ellipsis has been considered an antecedent of generative grammar.
Concerning L2 acquisition, the conceptualisation of 'native' versus 'non- native' language knowledge emerged. In L2 pedagogy, a key development was the incorporation of L1/L2 comparison and translation. In addition, usage- based instruction began to flourish alongside the traditional grammar- based instruction. A key interaction of the ideas of universal grammar and L2 acquisition occurred in the assertion by some early seventeenth-century language teachers that L2 learning should begin with the study of general language structure (i.e., that which is common to all languages), and then approach the specific L2. Finally, a further innovation towards the end of this period was the attempts by a number of scholars to create artificial languages intended for universal communication.
Chapter 7 begins with the work of late seventeenth century grammarians on 'general' or 'rational' grammar. The scholars of general grammar, beginning with Port-Royal grammarians (named after the area of Paris where they worked), aimed to discover what is common among all languages. In other words, they worked on universal grammar. Their work is known to some modern-day linguists through Chomsky's Cartesian Linguistics (1966), in which he argues that general grammar is the precursor of generative linguistics. A considerable section of this chapter is devoted to examination of Chomsky's claims and the heated debate they provoked. This is followed by details of the ideas of key seventeenth and eighteenth century grammarians and philosophers, some of whom firmly embraced an innatist theory of mind, while others rejected it and argued that all knowledge is acquired through observation mediated by the rationalist design of the mind.
L2 pedagogy and conceptualisation of L2 learning was heavily influenced by universalist and rationalist ideas about language during this time. Grammars were produced that provided instruction on the common features of languages, assuming that such information could facilitate L2 acquisition. The use of memorisation as an L2 learning tool declined.
In the nineteenth century, emphasis on universal grammar faded again, and the focus returned to data and the particulars of different languages. This trend was reflected in language teaching, where instruction on linguistic universals was abandoned, and a variety of techniques competed to fill the gap. These included grammar-translation, exposure to natural spoken language with little or no use of the L1, and memorisation. Observation of L1 acquisition also influenced thoughts on L2 learning at this time, but opinions were divided as to whether children learn solely by memorisation, or whether they are innately creative.
Chapter 8, the final chapter, covers the twentieth century. The chapter begins with details of the diverse ideas of early-twentieth century European linguists, including Ferdinand de Saussure, Otto Jespersen, and the scholars of the Prague School. The scene then moves to the US, with a description of American structuralism. The latter focused on capturing the details of different languages by means of scientific observation of empirical data. American structuralism has often been characterised as being 'anti-universalist' and 'anti-mentalist', in contrast to generative grammar. Thomas argues, however, that these characteristics have been overstated. For example, the influential structuralist Leonard Bloomfield believed that language universals may exist but that there was as yet insufficient data about the world's languages to discover them. His aim was to elevate linguistics to a science by means of working with observable data. He avoided 'mentalist' theorising about the nature of the mind because this could be too speculative and too far removed from the empirical facts.
In L2 acquisition, contrastive analysis, involving point-by-point comparison of the L1 and L2 grammars so as to identify where they differ, developed alongside structuralism. It is identified as the precursor to generative L2 acquisition research, and criticised on the grounds that it failed to account for all the difficulties L2 learners appear to encounter. Thomas points out that this criticism exaggerates the original claims of contrastive analysis, which was that the technique could shed light on some of portion of what learners know and do not know.
The final part of this chapter details the work that defines modern linguistic theory and L2 acquisition theory. Corder (1967) and Selinker (1972) are seen as pivotal documents in setting the scene for modern L2 acquisition research, by proposing that 'interlanguage' is inherently systematic, like any natural language. In linguistic theory, current work is dominated by universalism, in the forms of Noam Chomsky's theory of universal grammar and Joseph H. Greenberg's pursuit of language universals. Thomas concludes by pointing out that neither the idea that L2 knowledge is epistemologically equivalent to L1 knowledge, nor the focus on universals and rationalism, are novel in the history of linguistics. What late twentieth century adds to the story is its scientific approach.
A brief Afterword concludes the book. Here, Thomas responds to some criticisms of Thomas (1998), an article that anticipated the present book, including an objection by Gass, Fleck, Leder & Svediks (1998) that study of L2 acquisition research pre-dating the modern discipline does not contribute to current work. Thomas counters, citing Law (2003), that the goal in considering texts from a different era is to suspend the pursuit of utility, and seek reward in feeling 'stimulated and "restored" (Kaplan et al. 1993) in unanticipated, intangible ways by virtue of the shift in consciousness that an encounter with another culture's point of view affords' (p. 192). As a member of Thomas's target audience -- a second language acquisition researcher working within generative grammar, with no previous knowledge of pre-1960s L2 acquisition history -- I feel I can confirm the veracity of Thomas's description of the promised reward. I do not know how I will make use of my newly acquired knowledge of the history of universal grammar and second language acquisition, but I certainly feel better educated for having read it.
This book would clearly be at home on the reading list of courses in the history of linguistics. In addition, it would be a thought-provoking addition to courses on L2 acquisition research. It is well-written and painstakingly referenced. I noticed only three trivial typographical errors. Rather than criticism, I offer only a 'wish-list': it would have been useful to have a glossary of the Greek and Latin terms that crop up throughout the history (for the hurried reader who is too impatient to return to where each term first appeared); also, I would have been interested to see more actual examples from the historical texts (or their translations) referred to.
Chomsky, N. 1964. Cartesian Linguistics: A chapter in the history of rationalist thought. New York: Harper and Row.
Corder, S. P. 1967. The significance of learner's [sic] errors. International Review of Applied Linguistics 5: 161-170.
Gass, S. M., C. Fleck, N. Leder, & I. Svetics. 1998. Ahistoricity revisited: Does SLA have a history? Studies in Second Language Acquisition 20: 407-421.
Kaplan, S., L. V. Bardwell, & D. A. Slakter. 1993. The museum as a restorative environment. Environment and Behavior 25: 725-742.
Law, V. 2003. The history of linguistics in Europe: From Plato to 1600. Cambridge: Cambridge University Press.
Selinker, L. 1972. Interlanguage. International Review of Applied Linguistics 10: 209-231.
ABOUT THE REVIEWER:
ABOUT THE REVIEWER
The reviewer is currently a research fellow at the University of Newcastle Upon Tyne, and has recently completed her PhD in second language acquisition, investigating the roles of L1 transfer and UG in the L2 acquisition of Japanese quantifier scope by Chinese-, Korean-, and English- speaking learners.