Introduction XXVII recognition,sentence processing and,given a highly relevant visual context,even spoken language comprehension.Later,it goes on to discuss research where eye movements have proven informative for spoken language comprehension,mental models,and semantic memory,in situations where the visual display is completely blank.This latter collection of findings points to the spatial structure underlying language,possibly especially accessible through the use of eye-tracking. 2.4.5 Speaking for the wordless:Methods for studying the foundations of Cognitive linguistics in infants by Amanda Brandone,Roberta Michnick Golinkoff,Rachel Pulverman,Mandy J. Maguire,Kathy Hirsh-Pasek Shannon M.Pruden The authors describe the most useful methods to studying the incipient knowledge non- verbal infants have of the events and actions they observe.Habituation and intermodal preferential looking paradigms are discussed in detail.In the habituation paradigm,re- searchers utilize infants'preference for novelty.Infants are repeatedly shown a stimulus until their visual fixation to the display has decreased by a predetermined criterion.Novel test stimuli are then introduced with the expectation that if infants notice and find the changes to be interesting,they will watch the novel displays longer than the familiar ones. In the intermodal preferential looking paradigm(IPLP),infants are presented with a split- screen video presentation and acentral audio.The premise is that if children understand what they are hearing,they should spend more time looking to the display that matches the audio.After illustrating these methods,the authors describe research findings that inform the cognitive linguistics program. 2.4.6 Experimental Study of first and second language morphological processing by Kira Gor This chapter discusses the experimental paradigm used in research on the processing of inflectional morphology by first and second language speakers.It mainly focuses on the available data on regular and irregular verb processing in English and several languages with complex inflectional morphology,such as Russian,Italian,German,Norwegian, and Icelandic.The continuing debates between the proponents of the dual-and single- system approaches and some "hybrid"theories center around the claim that symbolic rule processing cannot be influenced by input-based linguistic probabilities or phonolog- ical similarity to the existing words belonging to the "neighbors"and "enemies"clusters. Accordingly,the paper addresses the ways in which the role of linguistic probabilities can be studied experimentally.It reviews the elicitation techniques aimed at triggering real and nonce verb generation,and how by controlling the input to second language learners,ex- perimenters are able to compare the native and non-native mechanisms of morphological processing
JB[v.20020404] Prn:12/04/2007; 9:53 F: HCP18IN.tex / p.6 (372-445) Introduction recognition, sentence processing and, given a highly relevant visual context, even spoken language comprehension. Later, it goes on to discuss research where eye movements have proven informative for spoken language comprehension, mental models, and semantic memory, in situations where the visual display is completely blank. This latter collection of findings points to the spatial structure underlying language, possibly especially accessible through the use of eye-tracking. .. Speaking for the wordless: Methods for studying the foundations of Cognitive linguistics in infants by Amanda Brandone, Roberta Michnick Golinkoff, Rachel Pulverman, Mandy J. Maguire, Kathy Hirsh-Pasek & Shannon M. Pruden The authors describe the most useful methods to studying the incipient knowledge nonverbal infants have of the events and actions they observe. Habituation and intermodal preferential looking paradigms are discussed in detail. In the habituation paradigm, researchers utilize infants’ preference for novelty. Infants are repeatedly shown a stimulus until their visual fixation to the display has decreased by a predetermined criterion. Novel test stimuli are then introduced with the expectation that if infants notice and find the changes to be interesting, they will watch the novel displays longer than the familiar ones. In the intermodal preferential looking paradigm (IPLP), infants are presented with a splitscreen video presentation and acentral audio. The premise is that if children understand what they are hearing, they should spend more time looking to the display that matches the audio. After illustrating these methods, the authors describe research findings that inform the cognitive linguistics program. .. Experimental Study of first and second language morphological processing by Kira Gor This chapter discusses the experimental paradigm used in research on the processing of inflectional morphology by first and second language speakers. It mainly focuses on the available data on regular and irregular verb processing in English and several languages with complex inflectional morphology, such as Russian, Italian, German, Norwegian, and Icelandic. The continuing debates between the proponents of the dual- and singlesystem approaches and some “hybrid” theories center around the claim that symbolic rule processing cannot be influenced by input-based linguistic probabilities or phonological similarity to the existing words belonging to the “neighbors”’ and “enemies”’ clusters. Accordingly, the paper addresses the ways in which the role of linguistic probabilities can be studied experimentally. It reviews the elicitation techniques aimed at triggering real and nonce verb generation, and how by controlling the input to second language learners, experimenters are able to compare the native and non-native mechanisms of morphological processing
xxvI The Editors 2.5 Neural approaches 2.5.1 Electrifying results:ERP data and cognitive linguistics by Seana Coulson Event-related brain potentials(ERPs)represent electrical activity in the brain that is time- locked to the onset of a cognitive or motor event.A direct index of brain activity with high temporal resolution,ERPs are known to be sensitive to many of the processing operations involved in the production and comprehension of language.But what are ERPs,really? And,what are they good for?This chapter reviews methods and data in the domain of electrophysiology of language comprehension.It begins with a general description of the electroencephalogram and event-related brain potentials(ERPs)and give an overview of language-sensitive ERP components such as the N400.Discussion of various ERP findings in psycholinguistics is intended to highlight how ERPs can be used to address questions about the representation and timing of cognitive processes,and how electrophysiologi- cal data can be used to complement experimental findings using behavioral paradigms. Finally,the author suggests how ERPs might be used to experimentally address issues in cognitive linguistics.It then describes the constraints one needs to consider in designing an ERP experiment,and gives examples of research which does and does not employ the strengths of the technique. 2.5.2 Bridging language with the rest of cognition:Computational,algorithmic and neurobiological issues and methods by Shimon Edelman The computational program for theoretical neuroscience proposed by Marr and Poggio calls for a study of biological information processing on several distinct levels of abstrac- tion.At each of these levels-computational(defining the problems and considering possible solutions),algorithmic(specifying the sequence of operations leading to a so- lution)and implementational-significant progress has been made in the understanding of cognition.In the past three decades,computational principles have been discovered that are common to a wide range of functions in perception (vision,hearing,olfaction)and action (motor control).More recently,these principles have been applied to the analysis of cognitive tasks that require dealing with structured information,such as visual scene understanding and analogical reasoning.Insofar as language relies on cognition-general principles and mechanisms,it should be possible to capitalize on the recent advances in the computational study of cognition by extending its methods to linguistics
JB[v.20020404] Prn:12/04/2007; 9:53 F: HCP18IN.tex / p.7 (445-471) The Editors . Neural approaches .. Electrifying results: ERP data and cognitive linguistics by Seana Coulson Event-related brain potentials (ERPs) represent electrical activity in the brain that is timelocked to the onset of a cognitive or motor event. A direct index of brain activity with high temporal resolution, ERPs are known to be sensitive to many of the processing operations involved in the production and comprehension of language. But what are ERPs, really? And, what are they good for? This chapter reviews methods and data in the domain of electrophysiology of language comprehension. It begins with a general description of the electroencephalogram and event-related brain potentials (ERPs) and give an overview of language-sensitive ERP components such as the N400. Discussion of various ERP findings in psycholinguistics is intended to highlight how ERPs can be used to address questions about the representation and timing of cognitive processes, and how electrophysiological data can be used to complement experimental findings using behavioral paradigms. Finally, the author suggests how ERPs might be used to experimentally address issues in cognitive linguistics. It then describes the constraints one needs to consider in designing an ERP experiment, and gives examples of research which does and does not employ the strengths of the technique. .. Bridging language with the rest of cognition: Computational, algorithmic and neurobiological issues and methods by Shimon Edelman The computational program for theoretical neuroscience proposed by Marr and Poggio calls for a study of biological information processing on several distinct levels of abstraction. At each of these levels – computational (defining the problems and considering possible solutions), algorithmic (specifying the sequence of operations leading to a solution) and implementational – significant progress has been made in the understanding of cognition. In the past three decades, computational principles have been discovered that are common to a wide range of functions in perception (vision, hearing, olfaction) and action (motor control). More recently, these principles have been applied to the analysis of cognitive tasks that require dealing with structured information, such as visual scene understanding and analogical reasoning. Insofar as language relies on cognition-general principles and mechanisms, it should be possible to capitalize on the recent advances in the computational study of cognition by extending its methods to linguistics
Foreword Leonard Talmy 1.Introduction The new insights into the system of conceptual structuring in language that have been coming from the relatively recent tradition of cognitive linguistics have rested mainly on the methodologies already standard in the field of linguistics overall:introspection in con- junction with theoretical analysis.The aim of the workshop that the present volume arises from was to help foster the application of additional methodologies to this emerging body of understanding.The spirit of the workshop and the papers here has been to value all of the applicable methodologies for their distinctive contribution to the total picture. Each methodology can be seen as having certain capacities and limitations that accord it a particular perspective on the nature of conceptual organization in language.In this respect,no single methodology is privileged over others or considered the gold standard of investigation. Though not all of them were represented at the workshop or are in this volume,the range of methodologies that apply to conceptual structure in language includes the follow- ing:introspection into the meanings and structures of linguistic forms and expressions, whether in isolation or in context,as well as the comparison of one's own introspec- tions with those reported by others (the more recent notion of "meta cognition"largely overlaps with that of introspection);the comparison of linguistic characteristics across typologically distinct languages and modalities (e.g.,spoken and signed language);the examination of how speech events interact with context,such as with the physical sur- roundings,the participants'background knowledge,or the cultural pattern;the analysis of audiovisual recordings of naturally occurring communication events,including their text,vocal dynamics,gesture,and body language;the(computer-aided)examination of collated corpora,often annotated;the examination of cumulatively recorded observations of linguistic behavior,as by children acquiring language;the experimental techniques of psycholinguistics;the instrumental probes of the brain's linguistic functioning in neuro- science;and the simulations of human linguistic behavior in artificial intelligence.Used in conjunction with all of these is the methodology of analytic thought,which includes the systematic manipulation of ideas,abstraction,comparison,and reasoning,and which is itself introspective in character,though with its object of attention not limited to lan- guage,as in the case of the linguistic introspection otherwise treated here.A selection of
JB[v.20020404] Prn:12/04/2007; 9:50 F: HCP18FR.tex / p.1 (47-114) Foreword Leonard Talmy . Introduction The new insights into the system of conceptual structuring in language that have been coming from the relatively recent tradition of cognitive linguistics have rested mainly on the methodologies already standard in the field of linguistics overall: introspection in conjunction with theoretical analysis. The aim of the workshop that the present volume arises from was to help foster the application of additional methodologies to this emerging body of understanding. The spirit of the workshop and the papers here has been to value all of the applicable methodologies for their distinctive contribution to the total picture. Each methodology can be seen as having certain capacities and limitations that accord it a particular perspective on the nature of conceptual organization in language. In this respect, no single methodology is privileged over others or considered the gold standard of investigation. Though not all of them were represented at the workshop or are in this volume, the range of methodologies that apply to conceptual structure in language includes the following: introspection into the meanings and structures of linguistic forms and expressions, whether in isolation or in context, as well as the comparison of one’s own introspections with those reported by others (the more recent notion of “meta cognition” largely overlaps with that of introspection); the comparison of linguistic characteristics across typologically distinct languages and modalities (e.g., spoken and signed language); the examination of how speech events interact with context, such as with the physical surroundings, the participants’ background knowledge, or the cultural pattern; the analysis of audiovisual recordings of naturally occurring communication events, including their text, vocal dynamics, gesture, and body language; the (computer-aided) examination of collated corpora, often annotated; the examination of cumulatively recorded observations of linguistic behavior, as by children acquiring language; the experimental techniques of psycholinguistics; the instrumental probes of the brain’s linguistic functioning in neuroscience; and the simulations of human linguistic behavior in artificial intelligence. Used in conjunction with all of these is the methodology of analytic thought, which includes the systematic manipulation of ideas, abstraction, comparison, and reasoning, and which is itself introspective in character, though with its object of attention not limited to language, as in the case of the linguistic introspection otherwise treated here. A selection of
XII Leonard Talmy these methodologies is considered next for their respective capacities and limitations,so as to demonstrate their complementary character. 2.Introspection In addition,introspection exhibits a particular profile of both capacities and limitations in accessing different aspects of language,as described below,and it is its pattern of limi- tations that has in part spurred the use of certain other methodologies to fill in for them The methodology of introspection begins this account and occupies some space because it has been central in the development of cognitive linguistics and continues as its main methodology,and because its particular profile of limitations has in part led to the pattern in the use of other methodologies.Linguistic introspection is conscious attention directed by a language user to particular aspects of language as manifest in her own cognition.More specifically,certain aspects of language spontaneously or through evocation can appear in a language user's consciousness-what can here be termed "first-level consciousness".In the same language user,a second level of consciousness can also occur that has as its ob- ject the contents of the first level of consciousness.This second-level consciousness-or attention-can be volitionally evoked and directed at a chosen linguistic target.Aspects of language differ in their readiness to appear in first-level consciousness.And,if present there,they differ in their amenability to second-level attention.An aspect of language is more amenable if it has greater strength and clarity and can remain more stably present in first-level consciousness while attention is directed at it,whereas it is less amenable if it is fainter,vaguer,or more elusive under such attempted scrutiny.As a cover term span- ning such first-level readiness and second-level amenability,aspects of language will here be said to differ in their "accessibility"to consciousness,attention,or introspection(see Talmy forthcoming). The accessibility of an aspect of language to directed conscious attention depends at least on the following five factors:cognitive organization in general,particulars of an indi- vidual's cognition,the current situation,conditions of attending,and the categorial object of attention.For the first factor,cognition across individuals appears to be structured in such a way-whether innately or from common developmental conditions-as to priv- ilege certain aspects of language over others along the accessibility gradient.As a second factor,though,due to individual differences in cognition-whether these result from in- nate differences or from training or practice-particular aspects of language can be above or below average in their accessibility to consciousness in a language user,thus able to diverge within limits from their usual ranking.As the third factor,the accessibility that various aspects of language afford to consciousness can vary over time within a single in- dividual in accord with changes in the situation or his concerns.An individual might,for example,attend more strongly to the exact wording of a lawyer and more to the tonality of an intimate.Fourth,the accessibility of various aspects of language to consciousness differs in accordance with what can be called the "conditions of attending",three of which can be suggested
JB[v.20020404] Prn:12/04/2007; 9:50 F: HCP18FR.tex / p.2 (114-187) Leonard Talmy these methodologies is considered next for their respective capacities and limitations, so as to demonstrate their complementary character. . Introspection In addition, introspection exhibits a particular profile of both capacities and limitations in accessing different aspects of language, as described below, and it is its pattern of limitations that has in part spurred the use of certain other methodologies to fill in for them. The methodology of introspection begins this account and occupies some space because it has been central in the development of cognitive linguistics and continues as its main methodology, and because its particular profile of limitations has in part led to the pattern in the use of other methodologies. Linguistic introspection is conscious attention directed by a language user to particular aspects of language as manifest in her own cognition. More specifically, certain aspects of language spontaneously or through evocation can appear in a language user’s consciousness – what can here be termed “first-level consciousness”. In the same language user, a second level of consciousness can also occur that has as its object the contents of the first level of consciousness. This second-level consciousness – or attention – can be volitionally evoked and directed at a chosen linguistic target. Aspects of language differ in their readiness to appear in first-level consciousness. And, if present there, they differ in their amenability to second-level attention. An aspect of language is more amenable if it has greater strength and clarity and can remain more stably present in first-level consciousness while attention is directed at it, whereas it is less amenable if it is fainter, vaguer, or more elusive under such attempted scrutiny. As a cover term spanning such first-level readiness and second-level amenability, aspects of language will here be said to differ in their “accessibility” to consciousness, attention, or introspection (see Talmy forthcoming). The accessibility of an aspect of language to directed conscious attention depends at least on the following five factors: cognitive organization in general, particulars of an individual’s cognition, the current situation, conditions of attending, and the categorial object of attention. For the first factor, cognition across individuals appears to be structured in such a way – whether innately or from common developmental conditions – as to privilege certain aspects of language over others along the accessibility gradient. As a second factor, though, due to individual differences in cognition – whether these result from innate differences or from training or practice – particular aspects of language can be above or below average in their accessibility to consciousness in a language user, thus able to diverge within limits from their usual ranking. As the third factor, the accessibility that various aspects of language afford to consciousness can vary over time within a single individual in accord with changes in the situation or his concerns. An individual might, for example, attend more strongly to the exact wording of a lawyer and more to the tonality of an intimate. Fourth, the accessibility of various aspects of language to consciousness differs in accordance with what can be called the “conditions of attending”, three of which can be suggested
Foreword xI In the first condition,an individual uses second-level consciousness to attend in isola- tion to some aspect of language within some small excerpted portion of discourse such as a word or sentence.In the second condition,an individual endeavors to attend with second- level consciousness to aspects of language that appear in first-level consciousness in the course of an ongoing communication event that she participates in as speaker or listener. The third condition rests on the assumption that some trace of first-level consciousness of some aspect of language can continue to be present for a short time,gradually fading away,directly following its evocation in some communication event.In some cases and to some degree,an individual can use second-level consciousness to attend to this brief perseveration of first-level consciousness to examine what its contents had been during their immediately preceding activation.The fifth factor,the category of the object of at- tention,can range from the meaning of a word,through the grammaticality of a sentence, to the intonation of one's speech.This factor is best considered in conjunction with the properties of the prior factors,as in what now follows. 2.1 First condition of attending To begin with the first condition of attending,human cognition seems organized in such a way that categories of language aspects differ in their accessibility to introspection in isolation.Of the most accessible categories-ones that are strong,clear,and stable as ob- jects of attention in isolation-the foremost is meaning:the conceptual content associated with linguistic representations.Not only is meaning the aspect of language that linguis- tic introspection is best at,but,in addition,introspection has the advantage over other methodologies in seemingly being the only one able to access it directly.Meaning is a con- sciousness phenomenon and,if it is to be taken on as a target of research,introspection- itself a process occurring in consciousness-is the relevant instrumentality able to reach its venue.Introspection accesses meaning of several types.One type is the meaning of an individual word.Access tends to be greater for an open-class word than for a closed-class word,and greater for a concrete meaning than for an abstract one.Thus,one can readily attend to the meaning of the open-class concrete word bucket,less so to the meaning of the open-class abstract word relation,and perhaps still less so to the meaning of the closed- class and abstract words not and with.Also stable under scrutiny in isolation is the overall meaning of a phrase or sentence.Likewise accessible is the derived meaning of an idiom or figure of speech,such as a metaphor,as distinguished from any literal reading such a form might represent.Comparably,as is often apparent to someone pausing to choose his phrasing carefully in writing,one can attend directly to the appropriateness or good fit of a word with respect to its meaning in a given context (or to its register in a given context as mentioned next). Somewhat less directly involved with meaning but still highly accessible to attention in isolation is the register of a linguistic form-that is,apart from its actual referent,its character as relatively more colloquial,learned,child-oriented,etc.and,hence,its appro- priateness for use in an informal,formal,or child-oriented situation,etc.Finally,now apart from meaning or its context,linguistic cognition seems structured in a way to per- mit relatively easy conscious access to the grammaticality of a phrase or sentence-that is
JB[v.20020404] Prn:12/04/2007; 9:50 F: HCP18FR.tex / p.3 (187-236) Foreword In the first condition, an individual uses second-level consciousness to attend in isolation to some aspect of language within some small excerpted portion of discourse such as a word or sentence. In the second condition, an individual endeavors to attend with secondlevel consciousness to aspects of language that appear in first-level consciousness in the course of an ongoing communication event that she participates in as speaker or listener. The third condition rests on the assumption that some trace of first-level consciousness of some aspect of language can continue to be present for a short time, gradually fading away, directly following its evocation in some communication event. In some cases and to some degree, an individual can use second-level consciousness to attend to this brief perseveration of first-level consciousness to examine what its contents had been during their immediately preceding activation. The fifth factor, the category of the object of attention, can range from the meaning of a word, through the grammaticality of a sentence, to the intonation of one’s speech. This factor is best considered in conjunction with the properties of the prior factors, as in what now follows. . First condition of attending To begin with the first condition of attending, human cognition seems organized in such a way that categories of language aspects differ in their accessibility to introspection in isolation. Of the most accessible categories – ones that are strong, clear, and stable as objects of attention in isolation – the foremost is meaning: the conceptual content associated with linguistic representations. Not only is meaning the aspect of language that linguistic introspection is best at, but, in addition, introspection has the advantage over other methodologies in seemingly being the only one able to access it directly. Meaning is a consciousness phenomenon and, if it is to be taken on as a target of research, introspection – itself a process occurring in consciousness – is the relevant instrumentality able to reach its venue. Introspection accesses meaning of several types. One type is the meaning of an individual word. Access tends to be greater for an open-class word than for a closed-class word, and greater for a concrete meaning than for an abstract one. Thus, one can readily attend to the meaning of the open-class concrete word bucket, less so to the meaning of the open-class abstract word relation, and perhaps still less so to the meaning of the closedclass and abstract words not and with. Also stable under scrutiny in isolation is the overall meaning of a phrase or sentence. Likewise accessible is the derived meaning of an idiom or figure of speech, such as a metaphor, as distinguished from any literal reading such a form might represent. Comparably, as is often apparent to someone pausing to choose his phrasing carefully in writing, one can attend directly to the appropriateness or good fit of a word with respect to its meaning in a given context (or to its register in a given context as mentioned next). Somewhat less directly involved with meaning but still highly accessible to attention in isolation is the register of a linguistic form – that is, apart from its actual referent, its character as relatively more colloquial, learned, child-oriented, etc. and, hence, its appropriateness for use in an informal, formal, or child-oriented situation, etc. Finally, now apart from meaning or its context, linguistic cognition seems structured in a way to permit relatively easy conscious access to the grammaticality of a phrase or sentence – that is