Papers reviewed and determined not to be word norm studies. Use the flag icon to report errors or suggest re-inclusion.
18265 papers
Various objections are raised against current practice in co-occurrence analysis. The use of Yule's coefficient Y is then advocated.
Three models for word frequency distributions, the lognormal law, the generalized inverse Gauss-Poisson law and the extended generalized Zipf's law are compared and evaluated with respect to goodness of fit and rationale. Application of these models to frequency distributions of a text, a corpus and morphological data reveals that no model can lay claim to exclusive validity, while inspection of the extrapolated theoretical vocabulary sizes raises doubts as to whether the urn scheme with independent trials is the correct underlying model for word frequency data. The role of morphology in shaping word frequency distributions is discussed, as well as parallelisms between vocabulary richness in literary studies and morphological productivity in linguistics.
Does gender affect reactions to violations of expected conversational behavior? This study examined ratings of interactants involved in interruptive exchanges. Audio recordings of two-person interactions that varied in gender composition but were identical in script features were rated by judges on several scales, including the degree to which participants were seen to be argumentative, rude, and assertive. Results showed that interrupter sex did not affect ratings even though interrupters were evaluated differently than those they interrupted. However, gender composition significantly affected two of three derived factors, disrespect and assertiveness, such that when a woman interrupted a man, the pair was rated significantly more disrespectful and assertive than either of the two same-sex pairs. Conversational interruptions that occur among mixed-sex pairs are often interpreted not merely as individual infractions but as an assault on the established power relations.
To date, no fully suitable data model for lexical databases has been proposed. As lexical databases have proliferated in multiple formats, there has been growing concern over the reusability of lexical resources. In this paper, we propose a model based on feature structures which overcomes most of the problems inherent in classical database models, and in particular enables accessing, manipulating or merging information structured in multiple ways. Because of their widespread use in the representation of linguistic information, the applicability of feature structures to lexical databases seems natural, although to our knowledge this has not yet been implemented. The use of feature structures in lexical databases also opens up the possibility of compatibility with computational lexicons.
The word senses in a published dictionary are a valuable resource for natural language processing and textual criticism alike. In order that they can be further exploited, their nature must be better understood. Lexicographers have always had to decide where to say a word has one sense, where two. The two studies described here look into their grounds for making distinctions. The first develops a classification scheme to describe the commonly occurring distinction types. The second examines the task of matching the usages of a word from a corpus with the senses a dictionary provides. Finally, a view of the ontological status of dictionary word senses is presented.
This paper describes a natural language generation system known as VINCI, which accepts as input a formal description of some subset of a natural language, and generates strings in the language. With the help of an attribute grammar formalism, the system can be used to simulate on a computer components of several current linguistic theories. The program, implemented in C, runs under a variety of operating systems, including UNIX, MS-DOS and VM/CMS. In this paper we consider not only the design of the system, but also some of its applications in linguistic modelling and second language acquisition research.
FrameBuilder, a tool for computational lexicography, makes possible the efficient creation of theoretically sound and internally consistent lexical/semantic definitions to a lexical database by a linguistically inexperienced user. The first section describes FrameBuilder as a frame generation component (lexicon builder) of a natural language processing (NLP) system Subsequent sections detail the linguistic theoretical foundation of FrameBuilder, and describe how this expertise is encoded into a system of reasoning-based heuristics for linguistic semantic analysis.
The purpose of the present study was to identify the physiological characteristics corresponding to three affects (fear, anger, and joy), which were elicited through real situations in a laboratory. The subjects were asked to rate their psychological responses using the Affect Rating Scales for each affective situation. Physiological indices (diastolic and systolic blood pressures, heart rate, respiration rate and frequency of galvanic skin response) were measured. The subjects' affects can be characterized by two functions obtained through discriminant analysis. One discriminant function separated positive from negative affects; the other set apart anger from the remaining affects.
Abstract Hierdie artikel het hoofsaaklik ten doel om die gedagte uit te lig dat vertalers nie werklik ejfektiewe vertalings sal kan lewer as hulle nie die (taal-)norme ken van die taal waarin hulle vertaal nie. Tale verskil t.o.v. veral vier aspekie, nl. die grammatika (strukture en patrone), die woordeskat, die styl en die kultuur (van die betrokke taalgroep) en dis belansrik dat elkeen van hierdie aspekte volledig in ag geneem moet word by 'n vertaling. Die primêre argument is dat 'n vertaler beter in staat sal wees om bogenoemde aspekte te verreken in sy vertaling as hy die norme (en daarmee saam die bronne waarin die norme uiteengesit word) ken en kan hanteer. The main aim of this article is to demonstrate that a translator will not be able to deliver a good and accurate translation if he is not familiar with the linguistic norms of the target language. Languages differ in at least four important aspects, i.e. grammar, vocabulary, stylistics and culture and these aspects certainly will have to be taken into consideration during the translation process. The primary argument is that a translator would be able to justify and also to motivate the changes he brings about if he knew the norms of the specific language (and also the reference sources explicating these norms).
Morphological information is useful for parsing, lemmatization, and in several natural language applications: text generation, machine translation, document retrieval, etc. In this paper, we shall be less concerned with what morphological processing systems are like (cf. Sproat 1992) than with the applications of computational morphology. We first present the kind of morphological information used by NLP (natural language processing) systems. That information is inflectional or derivational and may be encoded in lexical databases or retrieved dynamically through simple processing. The best known computational systems are presented, including some new methods to automatically acquire morphological information. Secondly, several technological applications using morphological information are described. In these sections, we have chosen to favor the description of some representative works in the corresponding fields, so as to illustrate how morphology is involved; we do not lay claim to exhaustiveness. This review aims at showing how truly useful morphology is for NLP systems.
Predictability and controllability of events influence attributions and affect in many research domains. In face-to-face social interaction, behavior is predictable from actor's own past behavior (internal determinants) and from partner's past behavior (social determinants). This study assessed how affect ratings are related to predictability of vocal activity from internal and social determinants. Time and frequency domain analysis of on-off vocal activity from 55 dyadic gettingacquainted conversations provided indexes of predictability from internal and social determinants. Greater predictability of vocal activity patterns from both internal and social determinants was associated with more positive affect. Future research should take internal as well as social determinants of behavior into account. The study of behavioral dialogues is emerging as an important research paradigm in social, developmental, and clinical psychology (Warner, 1991a). Investigators have examined time series data on the behavior, affect, or physiological states of social interaction partners to assess how social behavior is structured in time and how the behaviors of partners are interdependent.
MINDS (Mental Information Processing and Neuropsychological Diagnostic System) was developed with the goal of integrating a number of independent and stand-alone test programs that are used in the diagnosis of psychological and neuropsychological health. The system runs under MS-DOS. The shell program integrates subject information with data obtained through the use of the individual test programs. The current test battery comprises tasks on memory, attention, and motor performance; these tasks require the use of additional peripheral response devices, which are controlled via a multiple I/O interface card. Questionnaires are also included; they have been developed with the author language shell program MicroCAT. MINDS is programmed to allow easy integration of new tests. As an example, the Motor Planning Test is described. The equivalence of the computerized questionnaires with existing tests is also discussed.
The present paper summarizes the major methods and results of the multi-dimensional approach to genre variation. The approach combines the resources of computational tools, large text corpora, and multivariate statistical tools (such as factor analysis and cluster analysis). It has been used to address issues such as the relations among spoken and written genres in English, and the historical development of genres and styles. The approach has also been applied to other languages; in this regard it has been used to address broader theoretical issues, such as the extent to which genre, and style variation are comparable cross-linguistically, and the linguistic consequences of literacy.
Word sense disambiguation has been recognized as a major problem in natural language processing research for over forty years. Both quantitive and qualitative methods have been tried, but much of this work has been stymied by difficulties in acquiring appropriate lexical resources. The availability of this testing and training material has enabled us to develop quantitative disambiguation methods that achieve 92% accuracy in discriminating between two very distinct senses of a noun. In the training phase, we collect a number of instances of each sense of the polysemous noun. Then in the testing phase, we are given a new instance of the noun, and are asked to assign the instance to one of the senses. We attempt to answer this question by comparing the context of the unknown instance with contexts of known instances using a Bayesian argument that has been applied successfully in related tasks such as author identification and information retrieval. The proposed method is probably most appropriate for those aspects of sense disambiguation that are closest to the information retrieval task. In particular, the proposed method was designed to disambiguate senses that are usually associated with different topics.
Receptive vocabulary of Hispanic children in Miami was tested in both English and Spanish with complementary standardized tests, the Peabody Picture Vocabulary Test (PPVT-R) and the Test de Vocabulario en Imágenes Peabody (TVIP-H). 105 bilingual first graders, of middle to high socioeconomic status relative to national norms, were divided according to the language(s) spoken in their homes. Both groups, whether they spoke only Spanish in the home (OSH) or both English and Spanish in the home (ESH), performed near the mean of 100 in Spanish receptive vocabulary (TVIP-H means 97.0 and 96.5); in contrast, ESH group children scored more than 1 SD higher in English than OSH group children (PPVT-R means 88.0 and 69.7, respectively). It appears, therefore, that learning 2 languages at once does not harm receptive language development in the language of origin, while it does lay the groundwork for superior performance in the majority language. Furthermore, an analysis of translation equivalents, items shared by both tests, shows that a statistically significant portion of bilingual children's lexical knowledge does not overlap in their 2 languages and is therefore not reflected in single-language scores.
Czech description: Tento text kriticky zkoum! povahu jazykov! normy a kodifikace, přićemž odmítá přístup založený na statistice nebo průměrném jazykovém citu. Autor argumentuje, že jazyk není pouhou technickou pomůckou, ale složitým systémem, jehož kultivaci mají určovat "mistři" – významní spisovatelé, básníci a filosofové – nikoliv administrativní rozhodnutí nebo počítačové programy. Zvláštní pozornost je věnována vztahu mezi jazykem a myšlením; úpadek jazyka a jeho zjednodušování nevyhnutelně vede k povrchnosti myšlení. Na příkladu psaní slova "filosofie" (oproti "filozofie") autor demonstruje důležitost uchování etymologických vazeb na řecké kořeny, které nesou hlubší významové konotace. Text zdůrazňuje, že cílem jazykových úprav by nemělo být zjednodušení, ale posílení vyjadřovací schopnosti a jemnosti jazyka. Filosofové jsou zde prezentováni jako strážci jazykové hloubky, kteří mají právo rozhodovat o své terminologii nezávisle na oficiálních kodifikátorech. English description: This text critically examines the nature of linguistic norms and codification, rejecting approaches based on statistics or the average "language feel." The author argues that language is not merely a technical tool but a complex system whose cultivation should be directed by "masters"—prominent writers, poets, and philosophers—rather than administrative decisions or computer programs. Special attention is paid to the relationship between language and thought; the decline and simplification of language inevitably lead to superficial thinking. Using the spelling of "filosofie" (as opposed to "filozofie") as an example, the author demonstrates the importance of preserving etymological links to Greek roots, which carry deeper semantic connotations. The text emphasizes that the goal of language modification should not be simplification, but rather the enhancement of expressive power and nuance. Philosophers are presented as guardians of linguistic depth who must retain the right to determine their own terminology independently of official codifiers.
Nonhuman primates provide useful models for studying a variety of medical, biological, and behavioral topics. Four years of joystick-based automated testing of monkeys using the Language Research Center’s Computerized Test System (LRC-CTS) are examined to derive hints and principle for comparable testing with other species-including humans. The results of multiple parametric studies are reviewed, and reliability data are presented to reveal the surprises and pitfalls associated with video-task testing of performance.
The advancement of computer technologies, particularly the development of hypertext and interactive video, has presented to the academic community a new and effective tool for teaching and learning. An application of these technologies led to the concept of a hypermedia resource library—a set of integrated interactive computer modules that allow the user to browse and study topically specific content in a unique way. Such modules electronically present textual, graphic, and real-time video materials that instruct and quiz the user, offer a means for computer-based laboratory experimentation and data analysis, and provide statistical evaluation of the user’s progress. This paper will focus on the technology of computer-based hypermedia and the specific concept within this context of the artificially intelligent hypermedia resource library.
Proper Names (PNs) present a problem for the automatic processing and understanding of naturally occurring text. Due to their poor coverage in existing lexical resources and the continual appearance of new names, they represent a large body of unknown lexical data. Moreover, the complexity of the constructions in which they can appear and their own internal structure make them difficult to process, even if they are initially known. Yet the successful analysis of names is often crucial to the full understanding of a text. This paper proposes a solution to the problem and describes a natural language processing (NLP) system, FUMES, which makes use of the internal structure of names and the descriptive information that regularly accompanies them to produce lexical and knowledge base entries for unknown PNs. We present some preliminary results showing the viability of this approach for the identification of proper names.
The theory of motivated cheating postulates that test takers may cheat when they do not know an answer. With probabilityk, an “observer” is unsure of an answer and will copy from a nearby “target” with probabilityc. The corresponding parameters for the target may be entirely unrelated to those of the observer. Thus, the undesirable feature of bidirectionality of parameters found in correlational techniques is not an inherent feature of this theory of cheating. Predictions are derived, and estimates ofk andc are proposed. Statistically large values of c suggest that an observer was copying from a target. High values ofc for both the observer and the target suggest collusion. The theory is applied to a 40-item five-choice test taken by students in an introductory psychology section. From the full paired comparison matrix of target × observer parameter estimates, the method identifies 2 students who were probably in collusion.
Click to increase image sizeClick to decrease image size Notes1. I wish to thank Dr Lynn Williams of the University of Exeter for reading a draft of this article and making several suggestions.2. The Guernica Statute of Autonomy (1979) had effect in the three Spanish Basque provinces of Alava, Guipúzcoa and Vizcaya which became the three members of the Basque Autonomous Community. One of the aims of the 1982 Ley de normalización del uso del euskera is to protect every speaker's right to use Basque in the spheres of administration, education and in all means of communication. In Navarra, the fourth Spanish Basque province, the co-official status of Basque was recognized by the Parlamento Foral Navarro in November 1980.3. I follow Haugen's (1966) terminology here and interpret ‘vernacular’ as an underdeveloped language in the functional sense. E. Haugen, ‘Dialect, language, nation’, American Anthropologist, LXVIII (1966), 922–35.4. E. B. Ryan, ‘Why do Low-Prestige Language Varieties Persist?’, in H. Giles and R. N. St Clair (eds.), Language and Social Psychology, (Oxford: Basil Blackwell, 1979), 145–57.5. Pedro de Yrizar, Contribución a la dialectología de la lengua vasca, I (Zarauz: Caja de Ahorros Provincial de Guipúzcoa, 1981).6. Bonaparte made five trips to various parts of the Basque Country between 1856 and 1869. His extensive research allowed a detailed classification of the regional varieties; one of his major contributions was the Carte des sept provinces basques montrant la délimitation actuelle de l’Euskara et sa division en dialectes, sous-dialectes et variétés, completed in 1863 and published in London in 1866.7. Bonaparte, 98.8. K. Rotaetxe, ‘La norma vasca: codificación y desarrollo’, Revista española de lingüística, XVII, 2 (1987) 219–44.9. P. Lafitte, Grammaire basque (Navarro-labourdin littéraire) (Bayonne: 1944)10. See K. Rotaetxe, 228—30, for an account of the grammatical and orthographic reforms carried out in the process of standardization and the extent to which they eliminated the characteristics of vizcaíno.11. Rotaetxe, 240–43.12. J. I. Olabuénaga et al., La lucha del euskara en la Comunidad Autónoma Vasca (Vitoria: Servicio Central de Publicaciones del Gobierno Vasco, 1983). This work is a presentation of the results of the 1981 Census and of extensive surveys concerning language competence, use and attitudes carried out in the Autonomous Community at the beginning of the 1980s.13. K. Rotaetxe does not make a distinction here between learning Basque and learning Batua. It is probably true that in the case of most learners Batua is the norm adhered to, and this would explain why those living in Vizcaya encounter difficulties when attempting to put their newly acquired language to use. It should be remembered, however, that some establishments in Vizcaya teach a standard form of vizcaíno. It is quite possible that the low percentage of successful learners in Vizcaya includes precisely those speakers who have been educated in this standard vizcaíno.14. The higher success rate in Alava could seem rather surprising if it is remembered that the Basque spoken there is similar to that spoken in Vizcaya and is classified as a variety of vizcaíno: it could be argued that learners in Alava will be faced by the same problems of linguistic distance from Batua as those confronting their counterparts in Vizcaya. It is very likely that this is so for learners in those areas of Alava where Basque is spoken by the majority of the population. However, in Alava as a whole, the awareness of the distance between vizcaíno and Batua is not as acute as it is in Vizcaya, since the Basque-speakers form a very small minority and do not constitute a strong vizcaíno-speaking community. In Vitoria, capital of Alava and of the Basque Autonomous Community, there is a higher percentage of Basque-speakers, but they have migrated from several parts of the Basque Country and do not form a linguistically homogeneous group. The polarization vizcaíno/Batua does not, therefore, occur to the same degree.15. The grammatical calques on Castilian are not so much an intrinsic feature of Batua as a reflection of the fact that most people who write in Basque also write in Castilian, and tend to translate from Castilian when writing in Basque.16. Whereas Batua, therefore, is sometimes criticized for reproducing Castilian syntax, there are some lexical items which illustrate how Batua uses native Basque formations whilst the regional dialects use Castilian loan-words (for example eskribatu, ‘to write’ of vizcaíno from the Castilian escribir is translated as idatzi in Batua). Some native Basque speakers regard these features of the Batua lexicon as excessively purist, whereas others seem to interpret them as indications of their own linguistic inadequacy. It is worth remembering, however, that the extent of Castilian influence on native Basque-speakers’ vocabulary is possibly not as great as they themselves sometimes claim.17. Inventario de arquitectura rural alavesa (Vitoria: Diputación Foral de Alava, 1981).18. Inventario, 184.19. he informant was given the freedom to complete the forms of the Padrón either in Basque (Batua) or in Castilian. The question dealing with language competence was presented in the following way in Castilian: Conocimiento de euskara Señale con una X su nivel de comprensión, habla, lectura y escritura. 1. Nada 2. Con dificultad 3. Bien 20. The form of categorization adopted in the presentation of the Padrón results is designed to give each informant a nivel global de euskara. The informant is classified according to his own evaluation of his competence in the four skills of Comprehension, Speaking, Reading and Writing. The Classifications are: Euskaldunes alfabetizados: individuals who understand, speak, read and write Basque well. Euskaldunes parcialmente alfabetizados: individuals who understand and speak Basque well, but read and write the language with difficulty. Euskaldunes no alfabetizados: individuals who understand and speak Basque well, but are unable to read and write the language. Cuasi-euskaldunes alfabetizados: individuals who understand Basque well or with difficulty, speak Basque with difficulty, and read and write well or with difficulty. Cuasi-euskaldunes no alfabetizados: individuals who understand Basque well or with difficulty, speak Basque with difficulty, but are unable to read and write the language. Cuasi-euskaldunes pasivos: individuals who understand Basque well or with difficulty, but are unable to speak the language. Erdaldunes: individuals who are unable to understand or speak Basque. Although an attempt has been made to allow for the fact that there is not, in the case of every speaker, a constant reduction in levels of competence from Comprehension to Speaking, Speaking to Reading, and Reading to Writing (that is, there is a recognition that some individuals will be more competent in written skills than in oral skills), the seven groupings referred to do not appear to cover all the permutations provided by the informants’ evaluations of their competence in the four skills.21. The variety of words used to refer to the Basque language can lead to confusion. Euskara batua (or Batua) refers to the standard norm. In Castilian, non-standardized dialectal forms are often referred to as vasco or vascuence.22. J. I. Ruiz Olabuénaga, Atlas lingüístico vasco (Vitoria: Servicio Central de Publicaciones del Gobierno Vasco, 1984); J. I. Ruiz Olabuénaga et al., La lucha del euskara.23. The informant was given the freedom to complete the questionnaire either in Basque (Batua) or in Castilian. (All interviews, however, were conducted in Castilian.) The question dealing with language competence was presented in the following way in Castilian:24. These age-divisions were decided upon for two reasons. For purposes of comparison it was important to respect the age-cohorts used in the presentation of the results of the Padrón. At the same time the divisions attempt to take into account some of the historical factors which have affected the use and acquisition of Basque.25. When the interview was conducted in the absence of other Basque-speakers it is very unlikely that the informant felt that he was being tested: my very limited knowledge of Basque did not allow me to challenge his evaluations. However, the fact remains that it was probably more difficult for an informant to stretch the truth when confronted with the researcher than when allowed to complete the questionnaire in privacy.26. Olabuénaga et al., La lucha, 28.27. I am grateful to I. Agote (Política lingüística, Gobierno Vasco) for suggesting the possibility of adapting these evaluation methods to sociolinguistic surveys.28. J. L. M. Trim, Developing a Unit/Credit Scheme of Adult Language Learning (Oxford: Pergamon, 1980).29. M. Oskarsson, Approaches to Self-assessment in Foreign Language Learning (Oxford: Pergamon, 1980).30. When asked in my survey to state which was their first language, 11 of the 75 informants answered ‘Castilian’, 58 answered ‘Basque’ and six answered ‘Both Castilian and Basque’. Those who have Castilian as their native language are most likely to be adults who have moved to Aramayona from other parts of Spain. In some cases their offspring, although competent Basque speakers because of their exposure to the language at school, will claim Castilian to be their native tongue.31. Padrón municipal de habitantes de la Comunidad Autónoma de Euskadi 3. Educación y euskara (Vitoria: Instituto Vasco de Estadística [EUSTAT], 1988).32. It must be remembered that these tables have been compiled from questions which were originally presented to the informants in Basque and Castilian. See Notes 12 and 16 for the original terms used. Note that in the Padrón the levels of competence followed the order Nada, Con dificultad and Bien, whereas in my survey the order was the reverse (Fácilmente, Con dificultad and Nada). In order to ensure consistency, I have arranged the results of the Padrón in Table 7 in the order Bien, Con dificultad and Nada.
This paper reports on the history and development of a new undergraduate course teaching computing for humanities students at the University of Aberdeen, and assesses some new teaching approaches developed in the course. It is noted that teaching computing to humanities students appears to be viewed with suspicion by some Computer Science and Humanities Departments. The two camps seem to fear, for different reasons, that issues and practices important to their disciplines will be compromised or watered down. This paper describes an attempt to reverse any such attitudes on the part of staff and students and to take undergraduates considerably beyond mere word processing and computer literacy. Various methods and techniques used in the course are presented and their value assessed. The importance of using a consistent computer interface to helping students form a stable conceptual model of computers is considered. We reflect on the value of teaching more about Human Computer Interaction and Artificial Intelligence than is usual in Humanities Computing courses. A number of lessons are drawn from the course.
Lexical collocations have particular statistical distributions. We have developed a set of statistical techniques for retrieving and identifying collocations from large textual corpora. The techniques we developed are able to identify collocations of arbitrary length as well as flexible collocations. These techniques have been implemented in a lexicographic tool, Xtract, which is able to automatically acquire collocations with high retrieval performance. Xtract works in three stages. The first stage is based on a statistical technique for identifying word pairs involved in a syntactic relation. The words can appear in the text in any order and can be separated by an arbitrary number of other words. The second stage is based on a technique to extract n-word collocations (or n-grams) in a much simpler way than related methods. These collocations can involve closed class words such as particles and prepositions. A third stage is then applied to the output of stage one and applies parsing techniques to sentences involving a given word pair in order to identify the proper syntactic relation between the two words. A secondary effect of the third stage is to filter out a number of candidate collocations as irrelevant and thus produce higher quality output. In this paper we present an overview of Xtract and we describe several uses for Xtract and the knowledge it retrieves such as language generation and machine translation.
This retrospective on statistical analysis of literature in the first twenty-four years of Computers and the Humanities divides the essays under review into four groups: the philosophical, the statistical analyses of language, the statistical analyses of literary texts, and the statistical analyses of themes. It begins with the question: must valid statistical analysis of any literary text be based on a complete linguistic description of the language of the text? It summarizes and evaluates over forty essays, giving details on works discussed, sample sizes used, statistical methods applied, and quotations from the researchers. The essay ends with a polemical summary of what has been done and what the future holds. It emphasizes the importance of extended pre-computational stages of learning about language and discourse analysis; reading previous research, building on and challenging theory; and the use of carefully crafted, small databases to test specific questions.
Precision food pellets made from purified ingredients by two different manufacturers (P. J. Noyes Company, Inc. and Bio-Serv, Inc.), and advertised as dust-free and nutritionally complete, were compared as reinforcers of leverpressing in rats. Single schedules (a variable interval [VI] of 30 sec and a fixed ratio of 20) were used to compare ability to maintain moderate and high response rates, and concurrent VI schedules were used to assess preference. The two pellets maintained equivalent moderate and high response rates on the single schedules. In the concurrent schedules, 3 out of 4 rats showed a small preference for the Bio-Serv pellets. Although both pellets performed adequately in normal humidity, high humidity caused the Bio-Serv pellets to soften and occasionally to be crushed in the pellet dispenser. Their softness may have been the reason they were preferred. The two pellets can be regarded as generally equivalent, except in high humidity. The only experiments in which they might not be used interchangeably are those involving choices between different reinforcers.
Although the model of English pronunciation in Dutch schools is, and always has been, British English (commonly known as Received Pronunciation, RP), not only teachers, but also informed laymen notice that the pronunciation of learners seems to be more and more influenced by American English. An investigation into the nature and spread of this influence therefore seems in order. This paper discusses some of the preliminary results of a research project which aims to give an inventory and description of the influence of American English (General American, GA) on the pronunciation of 10 phonological variables, among which are /æ/ in words like classroom and wineglass, and flapped /t/ in words like pretty and meeting. A second aim of the project is to find out to which the degree the American and British varieties are attractive to our population. Therefore a number of listening tests were administered: - a preference test, in which subjects had to indicate which pronunciation of a lexical item they thought (a) best (i.e. confirm to the school norm) and (b) they would prefer to use themselves. - an identification test, in which subjects had to indicate whether an item was pronounced in RP or in GA. - a matched guise test consisting of 12 versions of the same story, read by 8 speakers, 4 of them in both varieties. A preliminary inventory shows that in roughly 25% of all the pronunciations of single lexical items (word list style) we can speak of an 'American-like' pronunciaton. The variables that are pronounced most frequently GA-like are flapped /t/ in little, /æ/ in classroom, /a/ in hockey and postvocalic /r/ in morning. It also appears that RP is still the preferred variety on both the preference tests, although this preference decreases slightly when asked which pronunciation they would prefer to use themselves. Roughly 65% of the items was correctly identified as being RP or GA. Finally, the matched guise test showed a significantly high rating of GA female voices on all factors except for the factor 'school-norm'. RP males and females scored relatively high on this factor as well as on 'social status', but dropped considerably on the 'activity' factor and remained below the GA voices on 'personal affect'.
BOOK NOTICES 391 (as in defining ship as 'every description of vessel used in navigation not propelled by oars'). This creates 'lexical illusions' for both drafters and interpreters, since the normal meaning and use of a word continue to impinge on and distort the stipulated meaning. Analyzing the tactics of drafters to avoid ambiguity introduced by multiple modification (in examples like charitable or social institutions), B focuses attention on the distinction between natural-language resolutions of ambiguity (by paragraphing or repetition ) and artificial means such as tabulation. While B claims that his intended audience is 'lawyers and others who have an interest in the operation of legislation' (7), it nevertheless seems to me that a fair amount of linguistic background is required to fully appreciate some of the technical aspects of this work. The book is clearly not intended as an introduction to legal language in the way that Mellinkoff's The Language of the Law (Little, Brown & Co.. 1963) is. Rather, it serves as an overview of the linguistics of drafting and interpretation. The book might have benefitted from some discussion of recent approaches to legal language which treat legal interpretation as a species of literary criticism (e.g. the work of Stanley Fish and others), especially in light of B's position that literary criticism is the 'nearest parallel' (3) to legal interpretation. Nevertheless, this is a book that will provide much useful information to linguists interested in statutory language. [Edwin Battistella, University of Alabama in Birmingham.) Gender variation in Dutch: A sociolinguistic study of Amsterdam speech. By Dédé Brouwer. Dordrecht & Providence: Foris, 1989. Pp. xv, 126. Paper $19.00. The aims of this book are twofold: first, B wants to provide an accurate description of the sociolinguistic variants used by men and women; second, she tries to gain insight into the factors that underlie this gender-related variation. To answer these questions, she studies the use of standard versus nonstandard speech by a carefully selected group of subjects in an Amsterdam neighborhood. B's thesis is that the social difference in the roles and positions of women and men are reflected in their respective use of language. The most consistent sociolinguistic findings about gender variation in language have been found at the phonetic level, and for that reason B concentrates on pronunciation in her study of Amsterdam speech. She criticizes older studies that assume biological factors to be the source of differences between female and male language usage. In addition, she points out difficulties that arise from classifying the profession of housewife, or from establishing a woman's social position in terms of the socioeconomic status of her husband. It follows that the misclassification of women argues for the need to reevaluate findings on differences between women's and men's speech. As far as sociolinguistic variants are concerned, B's conclusions may be roughly summarized as follows: first, women use a standard variety of Dutch more frequently than men do. Second, the two primarily independent variables B used, children and outside employment, both promote the use of the standard variety in women as well as in men. Also, secondary independent variables, such as level of education, social network, and lifestyle, seem to affect language use. Third, women do not have a positive attitude towards the use of coarse language. Finally, B finds a prevailing belief that the use of the standard language is more suitable for a woman than for a man. Beyond these findings. B's second aim is to use statistical data to explain why gender differences in language behavior exist; she stresses that men's language is not the norm, and that women's language is not deviant. B speculates that different social norms for women and men are crucial in accounting for sex differentiation in language, and thus that language behavior is accounted for by different ideas about one's own social roles. Gender variation in Dutch provides a fine example of the sociolinguistic method and does not assume an extensive linguistic background. It therefore has a variety of potential readers. B's style is clear and concise, and she offers convincing support for her arguments. Knowledge of Dutch is not necessary...
BOOK NOTICES 391 (as in defining ship as 'every description of vessel used in navigation not propelled by oars'). This creates 'lexical illusions' for both drafters and interpreters, since the normal meaning and use of a word continue to impinge on and distort the stipulated meaning. Analyzing the tactics of drafters to avoid ambiguity introduced by multiple modification (in examples like charitable or social institutions), B focuses attention on the distinction between natural-language resolutions of ambiguity (by paragraphing or repetition ) and artificial means such as tabulation. While B claims that his intended audience is 'lawyers and others who have an interest in the operation of legislation' (7), it nevertheless seems to me that a fair amount of linguistic background is required to fully appreciate some of the technical aspects of this work. The book is clearly not intended as an introduction to legal language in the way that Mellinkoff's The Language of the Law (Little, Brown & Co.. 1963) is. Rather, it serves as an overview of the linguistics of drafting and interpretation. The book might have benefitted from some discussion of recent approaches to legal language which treat legal interpretation as a species of literary criticism (e.g. the work of Stanley Fish and others), especially in light of B's position that literary criticism is the 'nearest parallel' (3) to legal interpretation. Nevertheless, this is a book that will provide much useful information to linguists interested in statutory language. [Edwin Battistella, University of Alabama in Birmingham.) Gender variation in Dutch: A sociolinguistic study of Amsterdam speech. By Dédé Brouwer. Dordrecht & Providence: Foris, 1989. Pp. xv, 126. Paper $19.00. The aims of this book are twofold: first, B wants to provide an accurate description of the sociolinguistic variants used by men and women; second, she tries to gain insight into the factors that underlie this gender-related variation. To answer these questions, she studies the use of standard versus nonstandard speech by a carefully selected group of subjects in an Amsterdam neighborhood. B's thesis is that the social difference in the roles and positions of women and men are reflected in their respective use of language. The most consistent sociolinguistic findings about gender variation in language have been found at the phonetic level, and for that reason B concentrates on pronunciation in her study of Amsterdam speech. She criticizes older studies that assume biological factors to be the source of differences between female and male language usage. In addition, she points out difficulties that arise from classifying the profession of housewife, or from establishing a woman's social position in terms of the socioeconomic status of her husband. It follows that the misclassification of women argues for the need to reevaluate findings on differences between women's and men's speech. As far as sociolinguistic variants are concerned, B's conclusions may be roughly summarized as follows: first, women use a standard variety of Dutch more frequently than men do. Second, the two primarily independent variables B used, children and outside employment, both promote the use of the standard variety in women as well as in men. Also, secondary independent variables, such as level of education, social network, and lifestyle, seem to affect language use. Third, women do not have a positive attitude towards the use of coarse language. Finally, B finds a prevailing belief that the use of the standard language is more suitable for a woman than for a man. Beyond these findings. B's second aim is to use statistical data to explain why gender differences in language behavior exist; she stresses that men's language is not the norm, and that women's language is not deviant. B speculates that different social norms for women and men are crucial in accounting for sex differentiation in language, and thus that language behavior is accounted for by different ideas about one's own social roles. Gender variation in Dutch provides a fine example of the sociolinguistic method and does not assume an extensive linguistic background. It therefore has a variety of potential readers. B's style is clear and concise, and she offers convincing support for her arguments. Knowledge of Dutch is not necessary...
This paper discusses an implemented program that automatically classifies verbs into those that describe only states of the world, such as to know, and those that describe events, such as to look. It works by exploiting the constraint between the syntactic environments in which a verb can occur and its meaning. The only input is on-line text. This demonstrates an important new technique for the automatic generation of lexical databases.
After 50 hours of small group participation during 9 weeks, 91 young adults rated each same-group member's conduct on SYMLOG's dimensions of dominance, friendliness, and task-orientedness. Earlier, they made similar ratings twice, several weeks apart, on separate measures of self-acceptance and acceptance of others. Individuals' mean SYMLOG dominance ratings by group peers correlated much more highly with aggregated ratings for self-acceptance (.83) than for other-acceptance (.02), while SYMLOG friendliness correlated more positively with acceptance of others (.85) than with self-acceptance (.05). Self-ratings yielded parallel, but weaker associations. After attenuation corrections, these divergent approaches to assessing the interpersonal domain's central dimensions yielded empirically equivalent results. Both methods provide measures relevant to small group processes.
Gender differences in mathematics test performance that favor males are rarely found on tests of computation or other mathematical symbol manipulations. These differences appear primarily in tests that are labeled as tests of “mathematical reasoning” and consist largely of word problems. The content of the word problems’ cover stories is a possible source of gender bias. Some have suggested that students are discouraged from solving problems for affective reasons when the content of the problem is sex-typed for the opposite sex; cognitive science research on problem-solving processes suggests that familiarity of content would be likely to affect problem-solving performance. To test these hypotheses, an experiment was conducted in which underlying mathematics problems were clothed in four different cover stories: masculine, feminine, neutral familiar, and neutral unfamiliar. No effect of sex-typing was found; there was a highly significant but small effect of familiarity. Ratings of problem characteristics were also collected, primarily to guide and confirm the realization of the design intentions.
This study is an investigation of the concordance among different sources of information on ratings of depression in patients with Alzheimer's disease (AD). Informants were 75 outpatients with diagnosed AD (22 with major depressive disorder), their caregivers, and geriatric clinicians. Clinicians rated each patient on the Hamilton Rating Scale for Depression on the basis of (a) the patient's report, (b) the caregiver's report, and (c) the clinician's evaluation. Overall, patients perceived themselves as less depressed than did caregivers or clinicians; for nondepressed patients, caregivers reported patients less depressed than clinicians reported; for depressed patients, no significant differences were obtained between caregiver or clinician. Certain items were more discrepant among informants and significantly differentiated depressed from nondepressed AD patients. Level of patient dementia did not affect ratings. These findings support the need for comprehensive, accurate assessment of depression in AD patients.
Scholars and educators have studied the effects of bilingualism on cognitive and linguistic abilities for almost a century. The question has generally focused on how a child with more than one language mentally organizes language, and on the repercussions of bilingualism on cognitive and linguistic development. Both these questions grew out of what might be called the monolingual-norm assumption: the belief that monolingualism is the cognitive-linguistic norm and that the child's cognitive system is fragile and designed to cope with only one language. The monolingual-norm assumption gave rise to the negative myths surrounding bilingualism: bilingualism has been blamed for cognitive, social, and emotional damage to children (see Hakuta, 1986, for review).
This thesis was concerned with an examination of psychological well-being, defined as consisting of negative affect, positive affect, and self-esteem. Two distinct approaches to well-being were identified (i.e. personality and social role). It was argued that well-being is likely determined by both personality and social role influences, and that these two areas might themselves be interrelated. Both cognitive and affective personality factors were included in this research (i.e. dysfunctional cognitions and affect intensity). Roles were examined at a general level (number of roles and their cognitive representation in terms of complexity) and at a specific level that included assessments of role stressfulness, importance, and satisfaction. Four studies designed to examine the independent and combined impact of these variables on well-being were conducted. Each study built on the previous both in terms of the dimension of well-being examined, and design complexity. Studies 1 and 2 were cross-sectional while Studies 3 and 4 were longitudinal.;Study 1 developed a set of social role stimuli. It was demonstrated that it is useful to consider a broad range of roles in well-being research. Furthermore, role evaluations of stressfulness predicted the negative affect dimension of well-being. Finally, dysfunctional cognitions were also related to negative affect. Study 2 extended these basic findings to the outcome variable of self-esteem. In addition, a curvilinear relationship between the number of roles and well-being was detected, but role complexity was unrelated to outcome.;Study 3 examined all three dimensions of well-being and included a second personality variable. Dysfunctional cognitions predicted negative affect and self-esteem but not positive affect. Affect intensity failed to predict well-being. Number of roles and role complexity did not predict well-being, and the role appraisals were similarly found to be poor predictors of outcome. There were few relationships between the two domains of personality and roles. Finally, Study 4 gathered daily affect ratings, and again dysfunctional cognitions predicted negative affect, while affect intensity predicted positive affect. Number of roles was related to positive affect.;Overall, personality was more strongly related to well-being than were role variables.
Since 1978 research in the development of software dedicated to the specific problems of historical research has been undertaken at the Max-Planck-Institute für Geschichte in Göttingen. From a background of practical experiences during these years, a concept of what an appropriate ‘workstation” for an historian would be has been derived. It stresses the necessity of three components: (a) software, derived from a detailed analysis of what differentiates information contained in historical sources from such present in current material, (b) databases which are as easily available as printed books and (c) knowledge bases which allow software and data bases to draw upon the information contained in historical reference works. A loose network of European research projects, dedicated to the realization of such a setup, is described.
Gender differences in mathematics test performance that favor males are rarely found on tests of computation or other mathematical symbol manipulations. They appear primarily in tests that are labeled as tests of mathematical reasoning and consist largely of word problems. The content of word problem cover stories is a possible source of gender bias. Some have suggested that students are discouraged from solving problems for affective reasons when the content of the problem is sex-typed for the opposite sex; cognitive science research on the problem solving processes suggests that familiarity of content would be likely to affect problem solving performance. To test these hypotheses, an experiment was conducted in which underlying mathematics problems were clothed in four different cover stories: masculine, feminine, neutral familiar and neutral unfamiliar. No effect of sex-typing was found; there was an highly significant but small effect of familiarity. Ratings of problem characteristics were also collected, primarily to guide and confirm the realization of the design intentions, and a number of interesting features of the rating results are discussed.
Abstract The roles of status and affect in the working relationships in the United States among 125 lower level managers and the 35 middle level managers to whom they reported was examined by using the Grasha-Ichiyama Psychological Size and Distance Scale (GIPSDS). The GIPSDS uses a circle-drawing task to determine perceptions of interpersonal status (psychological size) and affect (psychological distance) and a 22-item status-affect rating scale to identify the specific psychological dimensions underlying the drawings. Boss-subordinate exchanges also were measured by using Graen et al.'S (1982) Leader-Member Exchange Scale (LMX). Subordinates drew circles that estimated their psychological size or personal status, as larger than the circles their bosses drew for them; the bosses, in turn, drew circles that estimated their personal status as larger than the circles that their subordinates drew for them. Lower level managers' ratings of their personal knowledge and expertise were the best predictor of their psychological size, whereas ratings of personal dominance were the best predictor for middle level managers. Compared with their subordinates, middle level managers reported less psychological distance or more positive affect in those interactions. Rating scale items assessing emotional support in such interactions accounted for the largest share of unique variance (independent variance accounted for by a particular variable) in predicting LMX scores.
650 LANGUAGE, VOLUME 67, NUMBER 3 (1991) a general linguistic audience are mixed with articles of interest for the specialist reader only. However, all the articles fit into the (probably too general) frame 'Writing system and orthography '. An unfortunate feature of the book is the absence of addresses and affiliations of the contributors. The editors' full addresses are given as Peking, China, and Kleve, Germany. [Jürgen Tesak, University of Freiburg.] Studies on the languages and the verbal behavior of the Pennsylvania Germans II. (Zeitschrift für Dialektologie und Linguistik, Beiheft 64.) Ed. by Werner Enninger, Joachim Raith, and Karl-Heinz Wandt. Stuttgart: Franz Steiner, 1989. Pp. 170. Paper DM 54.00. This is the second volume of a collection of essays, the earlier volume, edited by Werner Enninger alone, appeared in 1986 in the same series. In the preface to Vol. I Enninger states that, 'although the linguistic varieties of the Pennsylvania Germans are among the best-documented and most widely explored languages of all ethnic minorities in the United States, the picture is far from complete' (ix). He observes further that Pennsylvania German offers a particularly good model for the study of such phenomena as substratum interference of LI and L2, of borrowing from L2 into Ll, and of language decay and death. There have been numerous earlier publications on Pennsylvania German, and some contributors to this volume have published widely on this subject elsewhere. The present volume, as well as a number of others, have received their impetus from the Essen Delaware Amish Project Team. in this collection of eleven papers, some are markedly weaker either in style or in content. Jonathan Watt's 'Ll interference in written L2: A comparison between the Pennsylvania German and Koine Greek situations' (103-15) contributes little of importance regarding Pennsylvania German. Rudolf Post's 'The lexicography of Palatinate German: Its relevance for Pennsylvania German research' (71-9) reads very much like an oral presentation. Heinz Kloss' death obviously precluded a more polished version of 'Sociolinguistic parallels between the Mennonite speakers of Pennsylvania German (or Pennsylfaanisch) and of Plautdietsch ' (117-24). Of the remaining articles, I found three particularly interesting: John Costello's 'Innovations increasing syntactic complexity in the native language of bilingual children from 5 to 10: The case for Pennsylvania German' (3-16); Marion Lois Huffines' 'The convergence of language death: The case of Pennsylvania German ' (17-28); and 'Rules of speaking and their mediation: The case of the Old Order Amish' (137-68). by Werner Enninger. John A. Hostetler, Joachim Raith, and Karl-Heinz Wandt. Costello demonstrates that, influenced by English (L2). the Pennsylvania German (Ll) of children shows an increased complexity in syntax rather than simplification. He finds that these new structures seem to become permanent in the children's Ll and explains this by the fact that linguistic variation in Pennsylvania German is considered normal by speakers, so parents don't correct any 'innovations'. Their main concern seems to be that children learn the dominant language, English, well. Huffines argues that the Pennsylvania German community has two linguistic norms for case usage, one among nonsectarians and another among the Amish and Mennonites. Among the nonsectarians the dative is being maintained, but Pennsylvania German itself is dying Among the sectarians the dative has been lost—not necessarily due to a convergence towards English, but rather from an internal development similar to the one in some Low German varieties. The article by Enninger et al. tries, from an ethnography -of-communication point of view, to shed light on the speaking rules in an Old Order Amish context, and to illustrate 'that speaking is rule-governed but subject to culture-specific constraints' (166). Apart from a few typographical errors and a line of text missing on p. 77. this is a well-edited volume which is interesting because of its varied subject matter. Judging by the rate ofpublishing on the topic and the fact that this area of investigation is far from exhausted, we can expect further collections edited by one or all of the present editors. [Wolfgang P. Ahrens, York University, Toronto.] Theorien vom Ursprung der Sprache. 2 vols. Ed. by Joachim Gessinger and Wolfert von Rahden. Berlin...
650 LANGUAGE, VOLUME 67, NUMBER 3 (1991) a general linguistic audience are mixed with articles of interest for the specialist reader only. However, all the articles fit into the (probably too general) frame 'Writing system and orthography '. An unfortunate feature of the book is the absence of addresses and affiliations of the contributors. The editors' full addresses are given as Peking, China, and Kleve, Germany. [Jürgen Tesak, University of Freiburg.] Studies on the languages and the verbal behavior of the Pennsylvania Germans II. (Zeitschrift für Dialektologie und Linguistik, Beiheft 64.) Ed. by Werner Enninger, Joachim Raith, and Karl-Heinz Wandt. Stuttgart: Franz Steiner, 1989. Pp. 170. Paper DM 54.00. This is the second volume of a collection of essays, the earlier volume, edited by Werner Enninger alone, appeared in 1986 in the same series. In the preface to Vol. I Enninger states that, 'although the linguistic varieties of the Pennsylvania Germans are among the best-documented and most widely explored languages of all ethnic minorities in the United States, the picture is far from complete' (ix). He observes further that Pennsylvania German offers a particularly good model for the study of such phenomena as substratum interference of LI and L2, of borrowing from L2 into Ll, and of language decay and death. There have been numerous earlier publications on Pennsylvania German, and some contributors to this volume have published widely on this subject elsewhere. The present volume, as well as a number of others, have received their impetus from the Essen Delaware Amish Project Team. in this collection of eleven papers, some are markedly weaker either in style or in content. Jonathan Watt's 'Ll interference in written L2: A comparison between the Pennsylvania German and Koine Greek situations' (103-15) contributes little of importance regarding Pennsylvania German. Rudolf Post's 'The lexicography of Palatinate German: Its relevance for Pennsylvania German research' (71-9) reads very much like an oral presentation. Heinz Kloss' death obviously precluded a more polished version of 'Sociolinguistic parallels between the Mennonite speakers of Pennsylvania German (or Pennsylfaanisch) and of Plautdietsch ' (117-24). Of the remaining articles, I found three particularly interesting: John Costello's 'Innovations increasing syntactic complexity in the native language of bilingual children from 5 to 10: The case for Pennsylvania German' (3-16); Marion Lois Huffines' 'The convergence of language death: The case of Pennsylvania German ' (17-28); and 'Rules of speaking and their mediation: The case of the Old Order Amish' (137-68). by Werner Enninger. John A. Hostetler, Joachim Raith, and Karl-Heinz Wandt. Costello demonstrates that, influenced by English (L2). the Pennsylvania German (Ll) of children shows an increased complexity in syntax rather than simplification. He finds that these new structures seem to become permanent in the children's Ll and explains this by the fact that linguistic variation in Pennsylvania German is considered normal by speakers, so parents don't correct any 'innovations'. Their main concern seems to be that children learn the dominant language, English, well. Huffines argues that the Pennsylvania German community has two linguistic norms for case usage, one among nonsectarians and another among the Amish and Mennonites. Among the nonsectarians the dative is being maintained, but Pennsylvania German itself is dying Among the sectarians the dative has been lost—not necessarily due to a convergence towards English, but rather from an internal development similar to the one in some Low German varieties. The article by Enninger et al. tries, from an ethnography -of-communication point of view, to shed light on the speaking rules in an Old Order Amish context, and to illustrate 'that speaking is rule-governed but subject to culture-specific constraints' (166). Apart from a few typographical errors and a line of text missing on p. 77. this is a well-edited volume which is interesting because of its varied subject matter. Judging by the rate ofpublishing on the topic and the fact that this area of investigation is far from exhausted, we can expect further collections edited by one or all of the present editors. [Wolfgang P. Ahrens, York University, Toronto.] Theorien vom Ursprung der Sprache. 2 vols. Ed. by Joachim Gessinger and Wolfert von Rahden. Berlin...
An associative lexical net whose weights are computed on the basis of the co-occurrences of words using Hebb's rule has been built. The co-occurrences of word pairs are determined by shifting a window over a large body of text. To estimate the associative response to a given stimulus word, the corresponding node is activated and its activity is propagated in the net. The proposed model assumes that words with high activities after propagation correspond to the associative responses of human subjects. These predictions have been tested and confirmed using the association norms collected by Russel and Jenkins.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">></ETX>
This paper offers a brief survey of some important developments in the use of computers in making dictionaries and lexicons. Making a dictionary involves collecting the data, sorting and lemmatizing, editing and printing. Five major types of machine-readable dictionaries have developed from these procedures: Machine Readable Lexicons of individual authors, Machine Readable Dictionaries with codes for linguistic information, Machine Dictionaries with selected information, and Lexical Databases with lexical information abstracted from machine-readable dictionaries. The second edition of the QED is a machine-readable dictionary with codes that may provide the basis for a diachronic lexical database.