Abstract and Keywords
In the Oxford Handbook of the History of Linguistics leading scholars from around the world explore and discuss the complex of interconnected approaches, skills, and tasks that has characterized the study of language for more than two-and-a-half millennia.
Keywords: linguistic historiography, linguistic history, linguistics in Antiquity, medieval linguistics, eighteenth-century linguistics, nineteenth-century linguistics, twentieth-century linguistics, history of ideas
The Oxford Handbook of the History of Linguistics offers comprehensive coverage of the history of linguistics in a single volume and will serve as an introduction to the understanding of countless topics within the history of linguistics. This project began immediately after I had completed The Western Classical Tradition in Linguistics (Allan 2010a), which contains pretty much all I wanted to write on that subject; but even on topics within the history of linguistics that I covered in that book, there are other perspectives to be presented and, on many matters, much greater expertise than mine to be tapped. In addition there are the non-western traditions to consider. So the present volume was conceived1 as a book that would make a significant contribution to the historiography of linguistics on a very wide range of topics. Thirty-four chapters, many covering a variety of issues, were commissioned from scholars who are expert in the field outlined in the title for each chapter. The size of the book necessarily favours concision over expansiveness, but there is a vast bibliography pointing to sources for further inquiry in all the fields covered in the book for readers wishing to pursue a special interest.
The readership envisaged for the handbook includes those already knowledgeable about the history of linguistics, but it is principally intended for students of linguistics and those (not necessarily professional linguists) with an interest in a history of investigations into language, language origins, the media through which language is delivered, and the purposes to which language is put. The book chronicles centuries of explanations for language structures, language meanings, and language use, as well as the history of some applications of linguistics.
The Oxford Handbook of the History of Linguistics is loosely organised into six thematically grouped parts. Part I contains four chapters which look at linguistic studies of the basics of human communication: the origins of language (Chapter 1), the invention of writing (Chapter 2), the nature of gesture (Chapter 3), and of sign languages (Chapter 4). Part II consists of five chapters that examine the history of the analysis and description of sound systems used in human languages. Chapter 5 looks at the development of phonetics from earliest times; Chapter 6 focuses on instrumental (p. 2) phonetics; Chapter 7 deals with the discovery of sound change laws in the nineteenth century; Chapter 8, the history of phonology; Chapter 9 chronicles the history of ideas about sound symbolism. Part III comprises three chapters dealing with non-western traditions. Chapter 10 is about linguistics in East Asia, with the focus on China, which has the richest history, but briefly commenting on linguistic events in Korea and Japan. Chapter 11 discourses on the superbly worked out and comprehensive linguistic tradition in Ancient India which influenced linguistic events in China, the Middle East, and, from the end of the eighteenth century, Europe. Chapter 12 is on Semitic and Afro-Asiatic linguistics. Part IV surveys the history of grammar and morphology in Europe and North America, proceeding from Ancient Greece and Rome (Chapter 13) to late antiquity and the Middle Ages (Chapter 14), the Renaissance and beyond (15), morphology throughout the ages (16), universal grammar from the medieval scholastics to Chomsky (17), American structuralism (18), Chomsky's contribution and legacy (19), European linguistics in the twentieth century (20), and the characteristics of functional and cognitive grammars (21). Part V comprises six chapters that survey lexicography (22), aspects of semantics (chapters 23–25), pragmatics (26), and text/discourse studies (27). Part VI offers histories of the application of linguistics to the comparison and classification of languages (Chapter 28), within the fields of social and cultural theory (29) and psychology and brain sciences (30); Chapter 31 deals with applications of linguistics in education and translation, Chapter 32 the use of computers, and 33 the development of corpora; finally, Chapter 34 is a history of the philosophy of linguistics that is something of a summary of many topics discussed earlier in this volume.
I'll now review the chapters in a little more detail, offering a handful of my own comments on the way, without precluding the much richer experience of reading each chapter for yourself. In Chapter 1, ‘The Origins and the Evolution of Language’, Salikoko S. Mufwene discusses the phylogenetic emergence of language and questions the probable time of the emergence of the ancestor, or ancestors, of modern language. Are individuals or populations the agents in the emergence of language? Mufwene reviews relationships between language and thought, language and the brain, language and our anatomical architecture, as well as the social conditions that favour oral communication.
In ‘The History of Writing as a History of Linguistics’, Chapter 2, Peter T. Daniels shows that the language constituents most accessible to conscious control (words, syllables, tone, etc.) are most likely to be represented in orthography. All writing systems utilize a sequence of symbols that reflects the temporal order of utterance. The earliest writing functioned as a kind of memory prompt and was to some extent pictographic; but when Sumerian script was adapted to the unrelated language Akkadian, it (mostly) used the sounds of the characters to spell out words phonetically. Similar patterns of development are found elsewhere.
In Chapter 3, Adam Kendon records the ‘History of the Study of Gesture’. Gesture is bodily action, usually integrated with speech during the process of communication, and was first described as an accompaniment to oratory. During the Renaissance, (p. 3) contact with exotic peoples led to a better appreciation of the communicative value of gesture, pantomime developed as a form of entertainment, and the mistaken idea arose that gesture is a universal language. Come the eighteenth century, gesture in sign language for the deaf was studied and taught. In the nineteenth century Wundt suggested that gesture is the first modality of language, an idea that persists today.
In ‘The History of Sign Language linguistics’, chapter 4, Bencie Woll explores the similarities and differences between signed and spoken languages and among sign languages. Sign languages are largely mutually unintelligible, and where iconicity is present, it does not produce identical lexicons: even the English-based sign languages of Ireland, the UK, and the USA are distinct. Sign languages are a semi-autonomous gestural medium that exploits three-dimensional space, simultaneous combinations of a location where the sign is articulated, a configuration and movement of the hand, and additional bodily gestures. Sign languages were ignored by linguists until the late twentieth century.
Part II opens with Michael K. C. MacMahon's ‘Orthography and the Early History of Phonetics’ (Chapter 5), reviewing the growth of phonetics in Europe, the Middle East, and Asia. It began when phonetic markers, syllabaries, and alphabets were invented; later, ‘correct’ pronunciation was of interest in the delivery of venerated texts (religious, poetic, etc.) and for orating. MacMahon contrasts ‘the acumen of phoneticians such as Pāṇini and Patañjali’ in Ancient India with the virtual absence of any significant work on phonetics in Europe before the Renaissance. Interest in reforming orthography inspired Martianus Capella (c.470 ce), the Icelander known as the ‘first grammarian’ (c.1135), Charles Butler (c.1633), the seventeenth-century inventors of ‘real character’ (c.1647–68), and Henry Sweet (c.1877) to phonetic analysis. Instrumental phonetics developed from primitive beginnings in the late eighteenth century, and flowered in the twentieth.
Deborah Loakes reviews its history in Chapter 6, ‘From IPA to PRAAT and Beyond’. The International Phonetic Alphabet was established in 1886 by a group of language teachers in Paris, based on an alphabet designed by Pitman, of shorthand fame, with Ellis, who had studied British dialects. By the end of that century, sound could be recorded. Early instruments for recording speech and articulatory gestures (e.g. linguagrams, palatography, filming, X-rays) were comparatively crude, but there was a great leap forward with the development of computer-aided systems such as PRAAT and EMU that allow for all sorts of articulatory and acoustic analyses of the segmental and prosodic characteristics of speech. Loakes gives examples of their use in dialect research and for forensic purposes. The nineteenth century saw great strides made in phonetics, concomitant with the study of relationships among Indo-European languages that led to the postulation of sound-change laws.
In Chapter 7, ‘Nineteenth-Century Study of Sound Change from Rask to Saussure’, Kate Burridge reports on the era when the study of language moved once again from description to explanation with the rise of the comparative method. It stemmed from the proposals of Sir William Jones, via Rask's identification of systematic sound correspondences improved upon by Germanicist Jacob Grimm with his notion of (p. 4) Kreislauf (‘rotation’), the exceptions to which were explained away by Grassmann and Verner. Neogrammarians demonstrated insightful interactions between the phonological and morphological levels of language (see also Chapter 28). Meanwhile, Bopp adopted an organic view of languages (they are born, flourish, and degenerate), while Schleicher pictured a tree as the model for family relationships among languages. To this period belong modern linguistic canons like the primacy of speech; recognition that sound change is phonetically conditioned and regular, permitting proto-forms to be reconstructed; and the aphorism that yesterday's syntax is today's morphology.
Harry van der Hulst in ‘Discoverers of the Phoneme’, Chapter 8, writes that the development of alphabetic systems presupposes some recognition of phonology. With the possible exception of Pāṇini, early grammarians in the Indian, Chinese, Greek, Roman, and Arabic traditions discussed speech sounds without explicit recognition of the phoneme. In the mid-1870s the Kazan school (Baudouin de Courtenay and Kruszewski) studied phonemic alternations (morphophonology). They influenced the Prague School (c.1928–39), who focused on what were later called the ‘distinctive features’ of phonemes. For them, as for de Courtenay, the phoneme was a psychological unit, but for Trubetzkoy it was an abstract functional unit of phonological opposition. Later, in America, Jakobson elaborated the theory of distinctive features, breaking with the idea that phonemes are the smallest elements of phonological structure. Generative phonology adopted distinctive features, with the modification that they are articulatory rather than abstract functional units.
Part II ends with Chapter 9, ‘A History of Sound Symbolism’, in which Margaret Magnus shows that although the arguments of the ‘naturalist’ Cratylus were dismissed by Socrates in Plato's Cratylus, the distribution of phonemes across semantic classes is not random as Hermogenes would hold because there are phonesthemes in every language (e.g. the onset to flash, flare, flame, flutter, flicker, and the rhyme in flutter, stutter, clutter). Humboldt recognized the existence of sound symbolism, as did Jespersen and Jakobson. However, if sound determines meaning, we should know what a word means just by hearing it, yet we don't. Locke sided with Hermogenes, but Leibniz pointed out there must be some rationale behind the forms of names. We don't normally think of Bloomfield as a romantic, but he wrote: ‘to study the coordination of certain sounds with certain meanings is to study language’ (1933: 27).
Part III is on non-western traditions. In Chapter 10, ‘East Asian Linguistics’, Karen Steffen Chung notes that the earliest linguistic records in China are fifth-millennium bce symbols on pottery that might be precursors in modern Chinese script; Oracle Bone Inscriptions (jiǎgǔwén) from as early as the fourteenth century bce certainly are. The Ěryǎ (second century bce) explains characters found in earlier texts. Xu Shen's Shuōwén Jiězì (121 ce) is China's first dictionary. In Chinese lexica, items are arranged by subject category and usually paired with a phonetically related definition, either a homophone or an alliterative or rhyming gloss; such entries are more useful as a prompt for the correct form of a word than a means to discover meaning. Because Chinese has very little morphological marking, syntacticians from the third century ce simply divided content words from function words. The most notable linguistic event in Korea was the invention in 1446 of Hangul to replace Chinese script.
(p. 5) In Chapter 11, ‘Linguistics in India’, Peter M. Scharf examines the tradition of linguistic analysis that grew up around the preservation of the oral language of the ancient Vedic hymns composed between 1900 and 1100 bce. In the first millennium bce commentators often used faulty etymology when interpreting terms in ritual liturgy (something similar happened later in Europe: see Chapters 13 and 24). From the sixth century bce, systematic analyses of phonetics, phonology, and prosody often reveal dialect variation. Some phoneticians recognized component features of segments as distinct from both articulatory processes and the segments themselves—which directly inspired feature analysis in twentieth-century Europe (see Chapter 8). In the early fourth century bce, drawing on existing work, Pāṇini composed a precise and fairly complete description of late Vedic Sanskrit. There are phonetic, morphological, syntactic, and semantic rules and conditions, rule ordering, metarules, lexical and phonological lists. Commentaries and developments continued through to the Middle Ages.
In Chapter 12, ‘From Semitic to Afro-Asiatic’, Edward Lipiński suggests that Proto-Semitic, a branch of Afro-Asiatic, probably originated in the present-day Sahara. Bilingual Semitic word lists date from the third millennium bce, and tabled equivalences between Sumerian and Akkadian phrases from the eighteenth century bce. The earliest extant Syriac grammar dates from the late seventh century; the earliest Arabic grammar, that of Sībawayhi, from the eighth century; Hebrew tri-radicalism was noted in the tenth century. In the nineteenth century, decipherment of Egyptian hieroglyphs, Akkadian cuneiform script, and South Arabian and Ugaritic inscriptions was followed by grammatical studies that had a great impact on the analysis and perception of Semitic languages, and research extended to Egyptian connections with Beja and the Ethio-Semitic relations with Cushitic languages of the Horn of Africa.
Part IV consists of nine chapters that narrate the history of grammar and morphology in Europe and North America. The story begins with Chapter 13, ‘From Plato to Priscian: Philosophy's Legacy to Grammar’ by Catherine Atherton and David Blank. In Ancient Greece, language study grew out of philosophy: language enables truth-bearing presentations of the internal and external world and is also a vehicle of persuasion and education. The basics for the parts of speech can be found in Plato and Aristotle, but it was the Stoics who noted regularities and irregularities indicating underlying rules of grammar and norms of behaviour governing the use of language. In the second century bce Aristarchus of Samothrace refers to all eight traditional parts of speech and to some of their subcategories; these were propagated in the Tekhnê grammatikê (attributed to Dionysius Thrax) which was a model for the pedagogical grammar of Donatus—a cornerstone of Latin instruction throughout the Middle Ages. Although very little original material has survived, the Stoics were a major influence on Varro, Apollonius, and Herodian, and—indirectly—their disciple Priscian2.
(p. 6) Donatus and Priscian figure in Anneli Luhtala's ‘Pedagogical Grammars before the Eighteenth Century’, Chapter 14. Luhtala identifies four types of word and paradigm grammars in late antiquity. Those, like Donatus' school grammar, cover the parts of speech and are often cast in question-and-answer form. Then there are those, like Priscian's Institutes, which closely examine one or more parts of speech. Parsing grammars, which also favour the question–answer method, go for exhaustive analysis of the grammatical properties of words on the model of Priscian's Partitiones Duodecim Versuum Aeneidos Principalium. Commentaries undertake critical discussion to flesh out the terse structure of the typical school grammar, enabling new ideas to be launched with the stamp of authority from what was commented upon. The mnemonic value of verse grammars made them popular from the twelfth to sixteenth centuries. A major novelty in the medieval period was the integration into pedagogical grammars of practical techniques for construing and constructing sentences.
Chapter 15, ‘Vernaculars and the Idea of a Standard Language’ by Andrew Linn, builds on the pedagogical theme of the previous chapter with the idea that underpinning language teaching is a presumption that the goal of learning is mastery of the standard vernacular language alongside, and later in place of, Latin. The oldest grammar of a vernacular is the Irish Auraicept na n-Éces ‘The Scholar's Primer’, perhaps composed in the seventh century, though the oldest surviving manuscript is c.1160. Grammars of Italian and Castilian appeared in the fifteenth century, and other languages followed. The Bible was translated into vernacular languages from the fifteenth century.
In Chapter 16, ‘Word-Based Morphology from Aristotle to Modern WP (Word and Paradigm Models)’, James P. Blevins identifies a tradition that has roots in Ancient Greek and Pāṇini. Both classify words into paradigms in which sub-word elements are inflections; in the western model, no unit intervenes between sound and word. Aristotle and the Alexandrians focused on the systematicity of inflectional morphology, where the Stoics and Varro emphasized the irregularities of derivational morphology. Even though the Neogrammarians recognized morphological roots in the late nineteenth century, they went little further than Priscian (sixth century); there was no advance until the rise of morphemic theory in the early twentieth century. Today, analysis is an interpretive process in which bundles of morphosyntactic features are spelled out by realization rules.
Jaap Maat, in ‘General or Universal Grammar from Plato to Chomsky’, Chapter 17, reminds us that Plato concluded, after systematically considering the relationships among language, knowledge, and the world, that investigating language is not a suitable method for investigating reality. Aristotle's logic provided a rigorous framework for the analysis of meaning and for studying the relationships between language, mind, and the world. He demonstrated that the structure of meaning will often differ from the structure of language. Most importantly, he pointed the way to universal grammar by pointing out that, though languages may differ, human minds and the world around us are (broadly speaking) the same for all humankind. Scholars in the Middle Ages, assuming that all languages are basically structured in the same way, sought to explain the grammatical categories found in Priscian using Aristotelian notions of universal (p. 7) concepts of reality. These Modistae proposed modes of signifying that specify features of the denotata for the different parts of speech. In 1587 Sanctius proposed his doctrina supplendi, an elaborative structure making explicit the meaning of a surface sentence;3 this idea recurs in the Port-Royal grammar of 1660 on the basis that it is present in the mind—a position later held by Chomsky.
In Chapter 18, ‘American Descriptivism (“Structuralism”)’, James P. Blevins says the 1940s–1950s saw the rise of a distinctive American school of linguistics that emphasized synchronic analysis. Focus shifted from description of languages to the investigation of methods, techniques, and inductive theories about languages. Bloomfield initiated a rigorous methodology and taxonomic representations of language structures that include abstract units as well as segmental material. Post-Bloomfieldians took morphemes to be composed of phonemes, but this was not sanctioned by Bloomfield; the travesty was mitigated by the introduction of morphophonemics. Bloomfield was agnostic about the place of meaning in linguistics, but Harris excluded it completely. It was Harris who defined grammar ‘as a set of instructions which generates the sentences of a language’ (1954b: 260), and Chomsky who shifted interest from language description to theory construction.
In ‘Noam Chomsky's Contribution to Linguistics: A Sketch’, Chapter 19, Robert Freidin recounts that Chomsky's work on generative grammar began with his honours thesis in 1949, revised in a master's thesis and developed in The Logical Structure of Linguistic Theory, written 1955–6 and amended for publication in 1975. Ever since, Chomsky has continuously revised the formulations and developed his theories of grammar. He has controversially linked his linguistic theory first to psychology and later to biology. From the start he was defining grammar as the recursive specification of a denumerable set of sentences along with their structural descriptions. The Minimalist Program proposes a simplified theory of grammar that presupposes language is a component of the human mind/brain with substantial biolinguistic innate content from Universal Grammar (UG).4
Chapter 20, by Giorgio Graffi, surveys ‘European Linguistics since Saussure’. Saussure's Cours de linguistique générale was reconstructed from student notes in 1916, revised in subsequent editions, and much later augmented by his own rediscovered notes (Engler 1967–74). For Saussure, a linguistic element is structurally and semantically defined in terms of its relations with other elements. Although the Prague School was influenced by Saussure's view of la langue as a system of reciprocal values, they also believed that language change is systematic and that at every stage language bears traces of earlier states—in other words, contra Saussure, diachrony is relevant to synchrony (an idea found in Paul 1880). The Prague School distinction between the physical phone and the phoneme as an element of a functional system distinguishes phonetics (p. 8) from phonology (though the insight had occasionally appeared earlier5). They emphasized that the functions of language must be accounted for within a grammar. Thereafter European linguistics was fragmented and, with the exception of functonalism, had little effect on later developments in syntactic or morphological theory.
In Chapter 21, ‘Functional and Cognitive Grammars’, Anna Siewierska recounts that, for functionalists (inspired by the Prague School), the communicative function of language structures grammar; for cognitivists, language reflects the way we conceptualize the world and so create meaning through language. Whereas Chomskyans treat grammar as autonomous, functionalists and cognitivists believe that cognitive and discoursal factors shape grammars, both diachronically and synchronically—though neither camp seeks support from psycholinguistic experimentation. Functional grammars are essentially process-oriented whereas cognitive grammars are construction-based. In cognitive grammars there is no sharp division between lexicon and syntax: the notion of construction subsumes all linguistic units irrespective of their size and complexity.
Part V is concerned with studies of meaning. In several traditions there was lexicography; semantics was primarily occupied with word meaning and, among etymologists, lexical networks. The Stoics recognized illocutionary types and, perhaps under Stoic influence, Apollonius Dyscolus identified the link between clause type, mood, and illocutionary force (see Chapter 34 and Allan 2010a: ch. 6). Otherwise the meaning of clauses was the province of logicians rather than grammarians until the late twentieth century. Also in the late twentieth century, ancient studies of rhetoric and oratory were supplanted by developments in pragmatics and discourse analysis.
Chapter 22, ‘Lexicography from Earliest Times to the Present’, by Patrick Hanks, begins by describing decisions a lexicographer must make when deciding on the character of an entry: the choice between polysemy and homonymy, how to accommodate nominal phrases, whether to include rare or specialized words, etymological detail, and/or citations; how should entries should be organized and what metalanguage used? Hanks recounts the history of lexicography and lexicographical method in China, India, Persia, Greece, Rome, the Semitic world, and Europe, then reviews the different functions of a dictionary: to be polyglot or monolingual; (supposedly) to act as a standardizing force for a language; to record language change; or to act as a resource for language learners.
Chapter 23, ‘The Logico-philosophical Tradition’, by Pieter A. M. Seuren, opens with an account of Plato's notion of ‘proposition’ implicit in the discussion of truth as correspondence between the idea of an entity and its (not-)being. For Aristotle, the truth of a proposition is the mental act of assigning properties to entities. Where Aristotle had a semantic notion of hupokeímenon ‘subject’ and katēgoroúmenon ‘predicate’, when these terms were used by Alexandrine grammarians they became syntactic constituents. The discourse relevance of such terms came to be recognized (p. 9) again in the later nineteenth century, and then—decades later—the Prague School reinterpreted them in terms of information structure. Aristotle recognized a ‘law of the excluded middle’ which was questioned in his lifetime by Eubulides, author of some famous paradoxes including the liar paradox and the sorites (‘heap’) problem. If Aristotle were right and if the morning star denotes Venus and the evening star also denotes Venus, then the morning star is the evening star would be necessarily true; but, as Frege pointed out, this is not so. It led him to separate Bedeutung ‘extension’ from Sinn ‘intension’. Russell's theory of descriptions led to questions about the nature of presupposition; Geach's rediscovery of Burleigh's donkey sentences led to the invention of Discourse Representation Theory, able to capture the incrementation of information in a system of logic.
In Chapter 24, ‘Lexical Semantics from Speculative Etymology to Structuralist Semantics’, Dirk Geeraerts dismisses the etymology in Plato's Cratylus and Isidore of Seville's Etymologiae as speculative guesswork and very often wrong. However, if we understand the Ancient etymologists to be exploring lexical relations rather than undertaking etymology in today's sense, their endeavours seem less absurd. The development of diachronic lexical semantics marks a shift from speculative etymology to comparative historical linguistics. In nineteenth-century Europe, study of the historical development of words and meanings gave rise to dictionaries that charted the development of language.
Chapter 25, ‘Post-structuralist and Cognitive Approaches to Meaning’, also by Geeraerts, continues the history with Katz and Fodor's introduction of componential analysis into generative grammar that combines ‘a structuralist method of analysis, a formalist system of description, and a mentalist conception of meaning’. Geeraerts reviews Wierzbicka's Natural Semantic Metalanguage, Jackendoff's Conceptual Semantics, Pustejovksy's Generative Lexicon, prototype semantics, and frame theory.
According to Jacob L. Mey in Chapter 26, ‘A Brief Sketch of the Historic Development of Pragmatics’, pragmatics studies acts of linguistic communication in particular social situations, wherein face maintenance is highly important. There is a communicative presumption that if a person is apparently trying to communicate, it is worthwhile to expend effort trying to understand them. The Sophists in Ancient Greece advocated language as a tool for action and not merely a vehicle for truth and falsity. Humboldt, Peirce, Vendryes, and Bühler all emphasized the importance of language in use—which determines truth values and implicatures. The Gricean maxims are practical applications of some of Aristotle's proposals to get an intended meaning across.
Part V ends with ‘Meaning in Texts and Contexts’ by Linda Waugh et al. This chapter focuses on the switch from the structuralist tradition of sentence-level grammar and concern with lexical meaning (continued in the Chomskyan paradigm) to a consideration of meaning in use within conversations and written texts. The switch was inspired by the ideas of Wittgenstein and speech act theorists, the Prague School functionalists, the work of Firthian contextualists under the banner of Systemic Functional Linguistics, and analysts of real conversations, their structures, and the flow of information within them. This was a coming together of ideas from linguistics, (p. 10) philosophy, sociology, anthropology, ethnomethodology, and later narratology, critical discourse analysis, and the like, so that the wider panoply of linguistic history has come to examine the interplay of speaker/writer, hearer/reader, their belief systems, and the contexts in which language is produced and construed.
Part VI is a historiography of comparative, typological, sociolinguistics, psycho-/neurolinguistics, computational, and corpus linguistics, translation, and the philosophy of linguistics.
In Chapter 28, ‘Comparative, Historical, and Typological Linguistics since the Eighteenth Century’, Kurt R. Jankowsky credits Francis Bacon with promulgating practical empiricism over philosophical speculation, and Leibniz with applying this to the study of language. In the late eighteenth century Pallas and Adelung compiled comparative vocabularies and Sir William Jones launched discussion of the relationship of Sanskrit and Persian to European languages. Humboldt thought that the comparison of languages should look beyond vocabulary to the world view of language speakers. Rask, Grimm, Grassmann, and Verner mapped sound changes showing that careful attention to facts reveals regularities and genealogical relationships among languages. Comparative linguistics went hand-in-hand with historical linguistics, which then spawned typological linguistics. The Schlegel brothers and Humboldt began to inquire into morphological typology in the first quarter of the nineteenth century; in the twentieth century Sapir took it up and Greenberg went much further.
In ‘Language, Culture, and Society’, Chapter 29, Ana Deumert says that sociocultural linguistics emerged in incremental steps. For Humboldt, language is the formative organ of thought, such that cognition and perception are embedded in the inner structure of languages, reflecting a multiplicity of ways of conceptualizing the world around us—an idea that recurs in the Sapir–Whorf hypothesis. Humboldt emphasized individual agency and creativity, but for Whitney, language was consensual within society, a matter taken up by Saussure and others6. Dialectology flourished in mid-nineteenth-century Europe and often encompassed social variation. From 1966, Labov's The Social Stratification of English in New York City became the model for sociolinguistic methodology.
In Chapter 30, ‘Language, the Mind, and the Brain’, Alan Garnham reports that the mind/body distinction was made in Plato's Phaedo. Aristotle referred to the cognitive mechanisms attendant on language, but neither these nor the crucial role of the brain in mental functioning were investigated until modern times. Brain trauma reveals that some language functions are localized in the left (primary language) hemisphere and some supported in the other hemisphere (usually the right). Brain imaging reveals brain activation in response to language, but there is no evidence at all of a language-dedicated part of the brain and thus no empirical evidence for Chomsky's ‘language organ’; furthermore, it has never been explained how such an organ could possibly have evolved. Chomsky also denies that language learning is like other cognitive (p. 11) learning systems; however, connectionist experiments and work in AI show that this is probably wrong. Psychologists working within lexical semantics find that meanings are not stored in the mind as abstract symbols but embedded in encyclopedic knowledge; and neurologists find that, for example, when action verbs are understood, areas of the brain that control movement are activated. In other words: cognition is dependent on the way the human body interacts with the rest of the world.
Chapter 31, ‘Translation: The Intertranslatability of Languages; Translation and Language Teaching’ by Kirsten Malmkjær, shows that functionality is an important part of language learning and understanding. It is uncertain that any two people understand the same thing even if they speak the same language; so although languages are intertranslatable, they are rarely ever precisely equivalent, with poetic and figurative language more difficult to translate than plain language. Translation is possible because meaning is a relation between the utterance and its context, and because humans share mental processes, somatic relations, and the ability to accommodate to the behaviours of others. Translation requires that readers of the target language text respond the same way as readers of the source language text. Malmkjær reports that translation exercises only aid the learning of languages if the learner is persuaded they are purposeful and rewarding.
In ‘Computational Linguistics’, Chapter 32, Graeme Hirst opens with ‘The field of computational linguistics (CL) has its origins in research on machine translation (MT) in the early days of computing in the post-Second World War 1940s.’ Except in places like the Cambridge Language Research Unit (founded 1954), assumptions about language structure were naïve. Bar-Hillel said that fully automatic high-quality translation was impossible because of the encyclopedic knowledge needed; this became the standard view from the 1960s. Determining the intent, goals, and plans of a human interlocutor became an important theme in computational linguistics in the mid-1970s. Schank's ‘conceptual dependency’ and then his ‘scripts’ sought to solve the problem for artificial intelligence and natural language understanding, which they almost did for tiny fragments of language. Computational linguistics rejects the need for formal syntax and semantics, except lexical meaning and frames, finding greater use for complex representations of encyclopedic knowledge. Google Translate functions on purely statistical criteria.
In Chapter 33, ‘The History of Corpus Linguistics’, Tony McEnery and Andrew Hardie tell us that corpus linguistics began with the compiling of frequency lists in the late nineteenth century. Field linguists and those studying language acquisition compile corpora, usually very small. Chomsky correctly claimed that no corpus can completely represent a language, but a carefully assembled corpus will contain a representative sample of the language. Corpora often show that reported native intuitions about grammaticality are false. The Brown Corpus assembled before 1964 was one of the first; such early corpora contained little speech data, but recently that has changed. With the increased power of computers and capacity of data storage media, corpora have grown in size so that, today, corpus linguistics has become the computer-assisted analysis of very large bodies of naturally occurring text. Automated (p. 12) tagging has been standard in most corpora since 1971. Corpora are a major resource for lexicographers, functionalists, cognitivists, psycholinguists, and computational linguists; they readily permit cross-language, diachronic, sociolinguistic, and stylistic comparisons to be made.
The 34th and final chapter, ‘Philosophy of Linguistics’ by Esa Itkonen, retraces, with a different perspective, some of the ground covered elsewhere in the volume. In addition there is a long discussion of Aristotle's syllogistic, which went almost unchallenged for 2,200 years. Whereas Aristotle was interested in language only as medium for logic, poetry, or rhetoric, the Stoics investigated language for its own sake, perhaps influencing the valuable contributions of Varro and Apollonius. Apollonius has a notion of underlying structure being transformed into surface structure, and also an idea very much like that of illocutionary force.7 In today's re-evaluation of the importance of diachronic linguistics as a means of explaining the connection between successive states of the language, Itkonen sees a return to beliefs similar to those of Schleicher and Paul in the nineteenth century.
That ends my brief overview of all the chapters in the Oxford Handbook of the History of Linguistics—with the caveat that the foregoing sketches barely scratch the surface of the treasure to be found in the pages of this volume.
The western tradition in linguistics is substantially sketched in my The Western Classical Tradition in Linguistics, 2nd edn (Allan 2010a). Rivals to that include A Short History of Linguistics (Robins 1997), Western Linguistics: An historical Introduction (Seuren 1998), and The History of Linguistics in Europe: From Plato to 1600 (Law 2003). Works that go beyond the western tradition include Universal History of Linguistics: India, China, Arabia, Europe (Itkonen 1991), Concise History of the Language Sciences: From the Sumerians to the Cognitivists (Koerner and Asher 1995) and the three-volume History of the Language Sciences (Auroux et al. 2000–2006). There are, of course, many other highly respected works referred to elsewhere in the Oxford Handbook of the History of Linguistics—which I confidently and unreservedly recommend to the reader.
Sunshine Coast, Australia
(1) Originally together with Keith Brown: see the Acknowledgements.