Factors governing the existence and distribution of sentential subjects in English
Stanley Dubinsky
Linguistics Program,
University of South Carolina
William D Davies
Linguistics Department
University of Iowa

This paper principally concerns the status and distribution of sentential subjects, as in (1), and to a lesser extent, the relation of (1) to the extraposition construction shown in (2).

(1) That John left early disappointed us.

(2) It disappointed us that John left early.

We will show that some long-standing assumptions about the distribution of non-nominal subjects (going back to Koster 1978 and Stowell 1981) are incorrect, and that the distribution of these is broader than previous analyses have acknowledged. We suggest that previous conclusions about this class of data were driven by facts whose explanation likely lies outside the realm of syntactic structure. More recent work in corpus studies and psycholinguistics appears to provide better explanations for the distribution of sentential subjects. In particular, better explanations seem to be available in terms of prosody and phrasal weight (Erdmann 1988) or processing factors, such as the "integration cost component" of Gibson's
(1998) Syntactic Prediction Locality Theory (SPLT).

While much of the debate about the status of sentential subjects occurred 20-30 years ago, the issue is still quite relevant, given the fact that a number of current syntax textbooks still present what turn out to be questionable analyses of these structures (see Culicover 1997, Haegeman and Guéron 1999, Lasnik 1999 (and 1995), and Radford 1997). What follows here then is an attempt to lay out previously established facts as well as some new ones, to call attention to long-standing misconceptions about sentential subjects, to distill the correct generalization for subject positions in English, and to point to potential sources of new data on the topic.

Attention: When, Where, and For Whom?
Susan Gass
Department of Linguistics and Germanic, Slavic, Asian and African Languages
Michigan State University

This presentation will deal with the role of focused attention as it relates to different language areas and different proficiency levels. Attention has been argued by Schmidt (2001) to be “necessary to understand virtually every aspect of second language acquisition.”

The major part of this paper will be a discussion of two empirical studies that address these issues. The first considers the different role that attention plays in second language learning depending on the particular aspect of language (syntax, morphosyntax, lexicon) and the proficiency level of the learners. The second study addresses the question of the relationship between input and interaction in second language learning.

On translation: Traditional ideas and contemporary theories
Eusebio V. Llácer Llorca
Department of English and Germanic Philology
Universitat de València, Spain

Speculation, analysis and evaluation of hypotheses on translation studies contribute to make translation an interdiscipline, fascinating in itself and instructive for other fields, for all knowledge develops with the interplay of different languages. Although the process of translation of a text in order to reach the reader is often quite a complex and often traceable one, we in fact know that the transference takes place.

But above all, this talk tries to be clear in its evaluative description, so as to offer the audience an informed critique and aims to encourage translators to persevere in their undertaking. After a brief historical introduction, the talk is divided into three parts that will try to answer different questions that might be posed by the non-specialist in order to better comprehend the main bases of translation.

In the first section, diverse periods of the 20th century are reviewed, which correspond to different questions that the field of translation studies faces, questions related to fundamental concepts for a proper understanding of ideas, concepts, methods and traductological applications. The second focuses on the most recent theories, models and solutions, based on important notions and problems for translation. Finally, I will review the main ideas of this work as well as offer some conclusions and perspectives about the future of theoretical, descriptive and applied studies.

The Discourse of Spirit Language -- Gibberish and Silence - in Anglo-Saxon Magic Charms
Leslie Arnovick
Department of English
University of British Columbia

Anglo-Saxon magic charms contain both instructions for actions to be performed and words to be spoken (or written). At the same time incantatory language is set forth in the charm texts, so also may silence be prescribed. This paper reexamines the concept of gibberish and silence, two semantically empty phenomena, in oral performance, identifying both as crucial components of Anglo-Saxon charm texts. One goal of the new historical-pragmatic linguistic methodology is to reconstruct the pragmatics of earlier language use, asking how an audience from a particular historical period may have interpreted an utterance within its discursive context. My task here is to ask of Anglo-Saxon charms, not only "who speaks?", but who "listens?" In this paper I will argue that for the viewers of the charm rite, the "gibberish" common to several Old English charms actually comprises utterances with a clear pragmatic function. Gibberish utterances work magic: they perform the single illocutionary job of the charm: to speak and thereby to cure or to heal or to protect or otherwise to direct the powers-that-be. Representing more than an absence of sound, silence assumes another pragmatic function in the performance of Anglo-Saxon magic charms. Instructions for keeping silent command nothing less than dynamic acts of non-verbal communication aimed at the spirit world, the meaning of which is cooperatively inferred by the audience of the ritual. A survey of a corpus of Anglo-Saxon charms reveals that not only does silence serve to foreground and frame the utterance of magic words (acting as a discourse marker), but also it evokes or initiates a supernatural "performance arena" in which mystical communication may take place. The paper concludes by asking why crucial communications in the charm often clothe themselves in the language of "unknowing."

Belle's body just caught the fit gnat: The comprehension of Northern Cities Shifted vowels by local speakers
Dennis R. Preston
Department of Linguistics and Germanic, Slavic, Asian and African Languages
Michigan State University

Sociophonetic perception experiments have established an interesting array of facts about respondent abilities to detect regional and ethnic identity (e.g., Clopper and Pisoni 2004, Graff, Labov and Harrris 1986, Plichta and Preston 2003, Preston 1989, Purnell, Idsardi, and Baugh 1999), about language ideologies and their interference with perception (e.g., Niedzielski 1999, Morris 2003), and about the influence of variation on comprehension (e.g., Labov and Ash 1997, Plichta and Rakerd 2003). These last studies, however, have usually focused on cross-dialectal settings, in which speakers from one area or group are tested on their ability to understand words (singly or embedded in phrases) pronounced by speakers from another area or group. Labov and Ash (1997) look at region (broadly) and a few local features (age, status, education, ethnicity), but there has been no work which involves a more finely-tuned selection of sociolinguistic variables and respondent ability to comprehend words pronounced in an emerging variety.

This study looks at age, region (including both area and the urban-rural distinction), sex, status, and ethnicity in the lower peninsula of Michigan and seeks to determine the contribution of those various factors to the degree of accurate comprehension of single words spoken by a number of respondents whose advanced degree of participation in the Northern Cities Shift (NCS) was determined acoustically. The words tested included all six NCS vowels and the four upgliding tense vowels. At least one token containing each vowel was played twice for each respondent, and the responses were judged correct or incorrect on the basis of the word they wrote (based only on the vowel; a response of 'pit' for the word 'bit,' for example, was considered correct).

Statistical analyses of the results show a decreased rate of correct identification for male, rural, older, lower-status, African American, and Appalachian immigrant respondents, although age and status were not selected as significant factors. The identity of the vowels themselves was also studied as a separate (significant) group, and both the order of misidentification (i, oh, e, ae, ^, o - from most often misidentified to least) as well as the identity of the wrong responses is discussed.

The amazing conclusion that people who speak a variety understand it better is reached, and I speculate about why certain vowels are more often misunderstood than others.

System and Region in the English of Low-SES African-American Schoolchildren
Anne Charity
Department of Linguistics
University of Pennsylvania

Labov et al. (1968) found that when given a sentence repetition test, African-American adolescent boys recast Standard American English (SAE) structures into forms that were more commonly used in African-American Vernacular English (AAVE). Labov interpolated from this evidence that African-American children have a receptive knowledge of SAE forms before they can reproduce the forms in SAE. Following Labov et. al (1968) and Baratz (1969), 5- to 8-year-old African-American children in New Orleans, LA, Washington, DC, and Cleveland, OH imitated the sentences of a story presented in SAE by teachers in a school setting. The 15 sentences included many items that were possible mismatches between the child’s vernacular and SAE. Afterwards, the children retold the story in their own words. Children’s use of SAE and AAVE features in both tasks was analyzed.

Usage rates for many AAVE phonological and grammatical forms correlated inversely with age and reading achievement, suggesting that as children have more contact with the school language, they acquire greater sensitivity to stigmatized features of AAVE, even those that have been attested to be frequent in more casual speech settings. Furthermore, children who produced more AAVE features during imitation also tended to use more AAVE features in story retelling . Quantitative and qualitative regional differences were also observed. Overall, higher rates of AAVE phonological features occurred in New Orleans than in Cleveland or Washington DC. Although usage of grammatical features was similar across cities, a higher degree of variation occurred in New Orleans. Variation was also found in the production of a yes/no question intonation contour. I hypothesize that intonation contours may be a feature that distinguishes African-Americans from whites, and may perhaps trigger language prejudice against African-American children, even if their grammar and phonology in the school setting include few AAVE features.

It is possible, especially for young speakers, that the observed regional differences may reflect varying local social norms for AAVE usage across formal and informal settings. Such differences have long been described anecdotally, but this is a first attempt at the empirical demonstration of the variation that is felt to occur in AAVE.

References:

Baratz, J. C. 1969. A bi-dialectal task for determining language proficiency in economically disadvantaged Negro Children. Child Development 40.

Labov, W., P. Cohen, C. Robins, & J. Lewis. 1968. The Non-Standard English of Black and Puerto Rican Speakers in New York City. Final report, Cooperative Research Project 3288. Philadelphia: U.S. Regional Survey

The Indefinites: Parallels between L1 and L2 Acquisition of English Articles
Heejeong Ko
Department of Linguistics & Philosophy
Massachusetts Institute of Technology

Investigation of parallels between adult second language (L2) and child first language (L1) acquisition has been at the center of intensive research in current acquisition studies. The shared assumption underlying this research program is that parallels between L2 and L1 acquisition can deepen our understanding of the general human ability to acquire language. Adult L2 data reveal the process of language acquisition uninfluenced by the concurrent cognitive growth of child L1 learners. Child L1 data reveal language acquisition patterns unaffected by L1-transfer. Parallels between adult L2 and child L1 data, thus, constitute strong evidence that the course of language acquisition is partly determined by genetically built-in linguistic principles, known as Universal Grammar (rather than by extra-linguistic factors like cognitive growth and L1-transfer). In this talk, I investigate such L1-L2 parallels in the domain of English article usage, and argue that t here is a close parallel between L1 and L2 acquisition of article semantics.

In particular, I discuss L2 acquisition of English articles by L1-Korean speakers. Since Korean lacks articles, L1-transfer can be ruled out as an explanation for L2-English article usage by Korean learners. On the basis of my experimental results, I argue that L2 learners’ article errors are not random, but systematically reflect the role of a semantic feature: partitivity (a partitive noun denotes an individual that is a member of a set introduced by previous discourse). Drawing from well-established observations in the literature, it is argued that partitivity also plays a significant role in L1 acquisition of English articles. The current findings on L1-L2 parallels in acquisition of articles imply that ‘the’ overuse by L1- and L2-learners is tied to a semantic factor rather than to children’s egocentricity. This study also shows that L2 learners’ errors that are not induced either by L1-transfer or by L2 input reflect learners’ systematic knowledge on semantic features in Universal Grammar.

Overgeneralized Causatives in L2 English and L2 Spanish: The Role of the L1
Mónica Cabrera
Linguistics Department
University of Southern California

Overgeneralized causatives (I come it closer so it won’t fall, You cried her) have been widely attested in child L1 acquisition of English (Bowerman 1982, among others). Previous studies have found that adult L2 learners also overgeneralize causatives in their L2 (Moore 1993; Montrul 1997). Given the parallel between child L1 and adult L2 data, and the fact that overgeneralized causatives are unacceptable in the L1, it has been argued that access to Universal Grammar, and not L1 transfer, is the motivation of these errors. However, in a subsequent study on the acquisition of causative structures with manner of motion, L1 transfer effects were found (Montrul 2001). L1 English speakers overgeneralized causatives in L2 Spanish (*El general marchó a los soldados al campamento), and L1 Spanish speakers incorrectly rejected these structures in L2 English (The general marched the soldiers to the camp).

Based on my experimental study on adult L1 English / L2 English and L1 Spanish / L2 English learners, I argue that the phenomenon of overgeneralization of causatives in L2 acquisition can be reduced to L1 transfer of different aspects of L1 knowledge. At earlier stages of L2 acquisition (beginner and intermediate proficiency), the constructional meaning of lexical causatives trigger the overgeneralization of causatives, in particular, with verbs encoding change of state or location. However, at the advanced stage, when recovering from overgeneralization, L1 lexical specific constraints come into play. The data suggest that L2 learners transfer L1 properties gradually, and not all at once, which has implications on the characterization of the L2 initial state.

References:
Bowerman, M. 1982. “Evaluating Competing Linguistic Models with Language Acquisition Data: Implications of Developmental Errors with Causative Verbs”. Quaderni di Semantica 3. 5-66.
Montrul, S. 1997. Transitivity alternations in second language acquisition: A cross-linguistic study of English, Spanish and Turkish. Ph.D. dissertation, McGill University
Montrul, S. 2001. “Agentive Verbs of Manner of Motion in Spanish and English as Second Languages”. Studies in Second Language Acquisition 23, 145-151.
Moore, M. 1993. Second language acquisition of lexically constrained transitivity alternations: acquisition of the causative alternation by second language learners of English. Ph.D. dissertation, University of South Carolina.

Native Language Influence on the Acquisition of L2 Semantics: Acquiring Aspect in English and Japanese
Alison Gabriele
Graduate Center
The City University of New York

Research on adult second language (L2) acquisition has only recently begun to explore questions of how meaning is acquired, examining whether learners assign target-like interpretations to sentences in a second language. Coppieters (1987) originally drew researchers’ attention to the fact that even extremely advanced learners, who have seemingly mastered the syntax of the second language, still have difficulty with certain semantic distinctions, such as the aspectual difference between the two past tenses in French. The difficulty was most evident in learners whose native language did not make the same semantic distinction. Coppieters’ study raises several interesting research questions that the present study addresses: Is aspect acquirable by adult second language learners? Secondly, is difficulty with the acquisition of aspect due to differences between the native language and the target language?

The present study focuses on the acquisition of aspect in a bi-directional study of L2 English and L2 Japanese learners. We focus on differences in the semantics of the form that marks the progressive in the two languages. The two forms, be+ing in English and te-iru in Japanese, interact differently with the lexical semantics of the verb to which they attach. In Japanese, achievement predicates such as arrive are incompatible with a progressive interpretation, allowing only for a resultative interpretation of V+ te-iru as in (1). Yet with accomplishment predicates, such as eat a cake in (2), a progressive reading is strongly preferred. In English, the progressive denotes an ongoing interpretation regardless of verb type.

(1) Hikoki-ga kuko-ni tui-te-iru
plane-NOM airport at arrive te-PRES
The plane (arrived and) is at the airport.
(2) Gakusei-ga keki-o tabe-te-iru
student-NOM cake-ACC eat te-PRES
The student is eating a cake.

Results of an interpretation task suggest that for both groups of learners, acquisition of an interpretation that is not present in the native language can be facilitated by robust and consistent input. Learners have less difficulty adding an interpretation to their grammar: learners of English have acquired the target-like ongoing interpretation of achievement verbs in the progressive and learners of Japanese have acquired the resultative interpretation of achievement verbs + te-iru. However, learners still have difficulty ruling out an interpretation that is available in the native language and not in the L2. Difficulty with repressing an interpretation available in the L1 remains even in a group of near-native speakers of English. It is possible that explicit information as to what is not possible in the L2 is necessary for acquisition to proceed in adult foreign language learners. These results provide further evidence for persistent native language influence in the acquisition of properties at the L2 syntax/semantics interface (cf. Sorace, 1999, 2003).

Textual Criticism and Frederick Klaeber’s Beowulf
Robert Fulk
Indiana University
Class of 1964 Chancellor's Professor of English

In its day, Klaeber’s Beowulf was regarded as relatively conservative in its treatment of the text. In the atmosphere of general textual conservatism that has subsequently come to prevail in Old English studies, however, Klaeber’s practices have come to seem, to many, quite liberal. This poses a notable dilemma for the editors of the forthcoming revised edition: should Klaeber’s textual practices be altered? The problem asks to be looked at from a variety of perspectives, taking into account Klaeber’s own evolving views on textual criticism, the bases for the current conservative bent in Old English textual studies, the insights afforded by recent larger textual theory, and the aims and uses of a critical edition of an Old English poem in the twenty-first century. Particular insights can be gleaned from an examination of manuscript images, revealing some of the remarkable problems involved in representing manuscript text in a print edition. Ultimately, it is probably impossible to formulate a rational set of editorial guidelines to cover all textual contingencies, but a set of general principles, to be adhered to wherever possible, is by no means pointless.

Expression and self-knowledge
Dorit Bar-On
Department of Philosophy
The University of North Carolina at Chapel Hill

Philosophers and psychologists alike have tried to explain a kind of basic self-knowledge human subjects appear to have that is articulated in self-ascriptions such as "I am feeling dizzy", "I am thinking about my lecture right now", "I'm scared of that dog", etc. (Such self-ascriptions are often called "avowals" in the literature.) It is commonly thought that we enjoy a certain authority or privilege when we avow present mental states, and that this is an important mark of the commonsense distinction between so-called conscious mental or psychological states, on the one hand, and purely bodily states, on the other. In my recent book (Speaking My Mind: Expression and Self-Knowledge, Clarendon Oxford Nov 2004) I develop a philosophical
account that tries to respect the commonsense view without falling into objectionable Cartesianism about the mind. On my account, what gives avowals a special status is the fact that they serve to express rather than merely report the states that are self-ascribed. Although avowals share their semantics with bodily self-reports, they differ from such reports in their pragmatics and epistemology. We can gain insight into basic self-knowledge, I argue, if we see avowals as continuous with
non-linguistic expressive behavior of the sort found in non-human animals and prelinguistic children. I take it as an advantage of my view that it lends itself to empirical support (or refutation) by studies in psychology and linguistics. In this talk, I aim to present the expressivist view I propose in a way that would invite suggestions regarding its potential empirical import.

Linguistics, Probability, and the Dating of Old English Verse Texts
Robert Fulk
Indiana University
Class of 1964 Chancellor's Professor of English

Historical linguists seem to be nearly unanimous in treating Beowulf and some other Old English poems as early compositions, and thus as reliable sources of data on the features of the most archaic stage in the history of English as a discrete language. This is particularly true in recent syntactic studies (e.g. by Pintzuk, van Kemenade, Y. Suzuki, and Hoch). Since there is no such agreement among non-linguists, and since the assumption has never been systematically defended in a linguistic context, it will be useful to assemble the data on archaic and innovative linguistic features, particularly in Beowulf, since these features do point to early composition, with remarkable consistency. An adequate understanding of the evidence, however, demands that some extralinguistic factors be taken into account, probably the most important of which is the way that the poetic register preserves archaic linguistic features long after their disappearance from everyday usage. Dealing adequately with such extralinguistic influences on the language of verse demands serious engagement with probabilistic reasoning, and this in turn suggests one way in which the much-discussed gap between linguistics and philology needs to be bridged.

Lost for words: Language impact on spatial cognition
Daniel Haun
Max Planck Institute for Psycholinguistics

The interaction between language and other cognitive functions is a ferociously discussed topic in the cognitive sciences. While some see cognition in its basic form as a set of innate concepts that simply find their communicative expression in language, others believe language to have a restructuring impact on the rest of cognition. One neglected aspect in this discussion is the investigation of the structure of language-independent cognition. In order to investigate the 'restructuring' viewpoint, we need to understand the initial structure.

In this experiment we compared great apes and 3-year-old children in a spatial memory task in which they display similar overall performance levels.

However, while apes prefer a place-based strategy, children outperform apes in feature-based spatial memory. Previous research with adult humans has reported language-dependence of feature-based memory in contrast to memory for location. Our data suggest that in non-linguistic tasks, children display an advantage over equally capable apes in strategies which are
impacted by language. This result is consistent with theories claiming that having a language can restructure other cognitive functions.

Convention, Conversation and Presupposition
Mandy Simons
Department of Philosophy, Carnegie Mellon University

In this talk, I will offer a novel account of the nature of presupposition. I will propose that presuppositions are properties, not of sentences, nor of speakers, but of utterances. Specifically, the presuppositions of an utterance are propositions which are required - in a sense to be specified - in order for the utterance to make sense to the addressee. I will spell out the notion of making sense in Gricean terms: an utterance makes sense to an interpreter just in case she can assign to it some interpretation which allows her to attribute full cooperativity to the speaker. The next step will be to articulate what it is for a proposition to be required in order to allow an interpreter to make sense of an utterance. I will argue that the relevant requirement concerns a somewhat complex relation which must hold between the proposition in question and the interpreter.

Having laid out the view of presupposition, I will show that many of the standard properties of presupposition are natural explicable in terms of the account, including the fact that presuppositions must generally be noncontroversial for the conversational participants.

last update: 4/01/05