Phonetics and phonology are among the branches of linguistics with least impact on applied linguistics. This is unfortunate, as they have a great deal to offer research and teaching in the many applications that investigate the production, understanding or representation of speech, especially second language teaching, which will be the focus of this chapter. One reason for their lack of impact might be they are often perceived as highly complicated topics, dominated by theoretical issues of limited relevance to practical applications. It is useful in this regard to invoke a distinction between 'complicated' and 'complex'. A simple system has few parts, related by a small enough number of rules as to be easily understood by the average person. A complicated system is quantitatively different, with many more parts, related by more numerous, more inter-related rules. A complex system is qualitatively different, with larger, less clearly-defined parts, connected by a smaller number of general, context-dependent principles (Ellis this vol). Working effectively with either kind of system requires recognition of which kind it is. However, since their products can seem superficially similar, it is possible to confuse them, with unfortunate results (Westley et al. 2006). The argument of this chapter is that speech is a complex system, but most current theories of phonetics and phonology model it as a complicated system. While this is appropriate for some applications, for others, a theoretical framework which recognises the complex nature of speech is needed. One problem is that understanding speech as a complex system means revising basic ideas in ways that challenge not just existing academic theories, but apparently obvious facts about speech. The intention here, however, is not to contradict existing ideas, but to place them in a wider context, with the aim of encouraging cross fertilisation between branches of theoretical and applied research that have had too little contact in recent decades. The chapter begins by reviewing some well-known observations, and equally well-known misconceptions, about speech. It then provides a simple analogy as a basis for understanding and comparing different views about speech, and goes on to use the analogy in an interpretive overview of the historical development of phonetics and phonology in relation to applied linguistics. Discussion then turns to how the knowledge acquired by phonetics and phonology can be framed in a way that allows fruitful, two-way interaction with various branches of applied linguistics, especially sociocognitive theories of second language teaching.

This paper speaks to a debate which has arisen across various branches of linguistics (see Eades 2009) regarding the relative levels of responsibility that should be given to (a) qualified linguists with professional expertise in a particular language, and (b) 'educated native speakers' of the language, in conducting Language Analysis for the Determination of the regional or social Origin of asylum seekers (LADO). It reviews existing evidence from the phonetics and sociolinguistic literature regarding the reliability of accent judgments by linguists and non-linguists. It argues that, while LADO is a valid form of assistance to offer in the asylum process, careful evaluation of its limitations, in general and in specific cases, is crucial, as in other branches of forensic linguistics. The paper concludes by calling for(a) a proper research program to investigate people's actual abilities in recognising, discriminating and identifying accents under various sociolinguistic conditions; (b) collaboration between LADO agencies and linguists to develop analysis and testing procedures; and (c) a system of accreditation by an independent, international authority for the agencies that carry out LADO.

Cambier-Langeveld (2010a) examines eight LADO case reports, attributing problems she finds in them to lack of native speaker competence on the part of the analysts, and uses the results to argue the legitimacy of a role for native speakers in LADO that is not currently allowed by the Guidelines (2004). The present article clarifies what the Guidelines say about analysts, and shows that the authors of the eight reports do not meet their requirements. It then argues that framing the important issues raised by Cambier-Langeveld in terms of a debate opposing 'native speakers' versus 'linguists' may not be the most helpful way forward, suggesting instead exploration of the question 'Under what conditions can LADO judgements be produced which are reliable enough for the human rights and national security issues at stake?'. It ends with a call for the Guidelines to be updated in light of recent work by Cambier-Langeveld and others, then administered effectively, preferably by an independent international body.

For many years, the difficulty adults have in learning second language pronunciation was explained with reference to the Critical Period Hypothesis-the belief that adults cannot, learn new phonological contrasts, perhaps due to physiological changes in the brain around puberty (Lenneberg 1967). This idea was in line with the contemporary theory of Generative Linguistics (Chomsky 1965), which took the view that language learning is achieved by an innate Language Acquisition Device which operates below the level of consciousness so is inaccessible to explicit teaching. It was also supported not just by the individual experience of many teachers (Macdonald 2002), but also by studies which appeared to demonstrate that explicit instruction in pronunciation was ineffective (Macdonald, Yule, and Powers 1994). For these reasons, pronunciation was given little attention in English language teaching during these years (Celce-Murcia, Brinton, and Goodwin 1996), and the belief that new contrasts could not be learned became a self-fulfilling prophecy, challenged -only by a few dedicated teachers (Baker 1981; Rogerson and Gilbert 1990).

One of the most basic aspects of learning a language is learning its words, or vocabulary. Applied linguistics rightly gives a good deal of attention to understanding the processes by which word meanings are learned, in both first and second language acquisition. But of course, words are not just meanings. Each word, as well as having a characteristic meaning, also has a characteristic sound, or pronunciation, which is just as important for learners to understand as its meaning. However, the processes by which the sounds of words are recognized tend to be given less attention in applied linguistics. This may be because recognizing spoken words seems so easy in everyday experience. The difficult issue, it seems, is recognizing written words, especially in languages like English that have irregular spelling. However, the reason spoken word recognition seems easy is because it is highly practiced and very familiar, not because it is simple - as becomes evident in second language contexts, when learners struggle to recognize even the simplest words (such as red or led). In fact, spoken word recognition is one of the most complex skills of human cognition, and the foundation of other crucial skills, especially (since words must be recognized before they can be reproduced) of pronunciation. Applied linguistics really needs a solid understanding of the complex nature of spoken word recognition, framed in a theory that offers practical guidance on how to respond to a variety of common problems in teaching and learning. This entry outlines some of the key findings of research on spoken word recognition and suggests how understanding this topic is relevant to applied linguistics, especially second language teaching.

This paper starts by recognising that, in general, pronunciation is the least successfully taught of the second language skills, and suggesting this indicates a need for a better theoretical framework within which teachers can understand and facilitate learners' acquisition of L2 pronunciation. Structural-generative theory, which has been dominant in phonology for some time, has limited application in this domain. However, applying the principles of Cognitive Phonology may lead to improved results. It then reviews the basic Cognitive Phonology principle: 'the signifier is a concept', and explains how the literacy bias (the tendency of those literate in the alphabetic script to believe that speech is a string of discrete phonemes) makes this principle more difficult to grasp than the very similar but far more widely understood principle that the signified is a concept. Discussion continues to consider implications of this idea for language teachers: phonemes, and other units of phonology, are not real things but abstract concepts. Teaching pronunciation thus involves facilitating concept formation. The paper then moves to consider some implications for theory of the observation that the concept of phoneme is derived from prior understanding of words and other larger units of phonology. It concludes by suggesting there may be productive parallels between the arguments presented here regarding the relationship between words and phonemes, and arguments advanced by Construction Grammar in regard to the relationship between lexis and grammar, whose implications for second language teaching are explored by other papers in this volume.

Language analysis for the determination of origin (LADO) is a relatively new branch of applied linguistics. Generally considered a forensic application, it is used by governments to assess asylum seekers applying for refugee status. The United Nations' 1951 Geneva Convention defines a refugee as a person with a well-founded fear of being persecuted in their home country for reasons of race, religion, nationality, membership of a particular social group, or political opinion. Asylum seekers who arrive in a country which signed the Geneva Convention, and are able to demonstrate such a well-founded fear, can claim the status of refugee, giving them rights of immigration to that country. Since many asylum seekers, for a variety of reasons, arrive without reliable identity papers, the first task of immigration authorities is to determine whether applicants' claims are valid, or whether they may be misrepresenting their identity in order to gain entry to the country. To aid in this determination, applicants are interviewed to test their knowledge of the language, culture, and region of the persecuted group they claim to belong to (European Union, 2008). In some cases, their speech is analyzed to determine whether it shows features of the specific language variety of the claimed group. This last procedure is LADO.

Many disciplines describe themselves as studying 'communication'. However observation of interdisciplinary discussion suggests that 'communication' may be conceptualized in different ways by different disciplines. This paper aims to promote mutual understanding among disciplines, not by proposing a universally valid definition of communication to which all disciplines should subscribe, but by, first, offering a set of questions that can be used to help disciplinary groups communicate their own views on communication to colleagues from other disciplines, and then creating a (preliminary) typology to map out the range of possible positions that can be taken in relation to those questions. Noting that academic disciplines have distinct cultures, the paper presents some concepts of intercultural communication as understood in applied linguistics that may be useful in facilitating interdisciplinary communication about communication.

The idea of the pronunciation teacher acting, not merely as a giver of information, but as a coach, is welcomed by many teachers. However, difficulty can be experienced with one of the key responsibilities of a coach, that of offering 'cues, suggestions and constructive feedback about performance' (Morley, 1991). It can be frustrating for both parties when learners find it hard to act on teachers' feedback to improve their pronunciation. This paper explores one factor in this difficulty: lack of shared metalanguage with which teachers and students can communicate about pronunciation. The term 'metalanguage' is often assumed to refer only to technical terminology for the phonetic description of speech. However, such technical metalanguage can be difficult to master - even for teachers, let alone second language learners - so is not always ideal for effective communication. This paper suggests a broader use of the term 'metalanguage', to refer to any way of speaking about speech and pronunciation, and advocates development of communicative metalanguage - consistent with, but complementary to, the technical metalanguage of phonetics and phonology - to suit the communication needs of different teaching contexts. The paper starts by reviewing some widely accepted insights about language and communication (Taylor, 2002), then shows how applying these same insights to pronunciation can enable teachers to provide feedback learners can understand and act upon easily. An important focus is the need for teachers to overcome the 'literacy bias', and recognise that teaching pronunciation involves helping learners with concept formation.

This article considers the reliability of transcripts used as evidence in court, especially transcripts of poor recordings. Background information about human speech and speech perception is presented, and the implications of this information for the use of transcripts of different kinds in legal contexts considered. Finally recommendations are made to allow judgement of the reliability of existing transcripts, ensure that newly created transcripts are reliable, and ensure that transcripts are presented to a jury appropriately.