Publications

Abstract

This paper presents a theory of the syntactic aspects of human sentence production. An important characteristic of unprepared speech is that overt pronunciation of a sentence can be initiated before the speaker has completely worked out the meaning content he or she is going to express in that sentence. Apparently, the speaker is able to build up a syntactically coherent utterance out of a series of syntactic fragments each rendering a new part of the meaning content. This incremental, left-to-right mode of sentence production is the central capability of the proposed Incremental Procedural Grammar (IPG). Certain other properties of spontaneous speech, as derivable from speech errors, hesitations, self-repairs, and language pathology, are accounted for as well. The psychological plausibility thus gained by the grammar appears compatible with a satisfactory level of linguistic plausibility in that sentences receive structural descriptions which are in line with current theories of grammar. More importantly, an explanation for the existence of configurational conditions on transformations and other linguistics rules is proposed. The basic design feature of IPG which gives rise to these psychologically and linguistically desirable properties, is the “Procedures + Stack” concept. Sentences are built not by a central constructing agency which overlooks the whole process but by a team of syntactic procedures (modules) which work-in parallel-on small parts of the sentence, have only a limited overview, and whose sole communication channel is a stock. IPG covers object complement constructions, interrogatives, and word order in main and subordinate clauses. It handles unbounded dependencies, cross-serial dependencies and coordination phenomena such as gapping and conjunction reduction. It is also capable of generating self-repairs and elliptical answers to questions. IPG has been implemented as an incremental Dutch sentence generator written in LISP.

Abstract

We present an overview of several corpus studies we carried out into the frequencies of argument NP orderings in the midfield of subordinate and main clauses of German. Comparing the corpus frequencies with grammaticality ratings published by Keller’s (2000), we observe a “grammaticality–frequency gap”: Quite a few argument orderings with zero corpus frequency are nevertheless assigned medium–range grammaticality ratings. We propose an explanation in terms of a two-factor theory. First, we hypothesize that the grammatical induction component needs a sufficient number of exposures to a syntactic pattern to incorporate it into its repertoire of more or less stable rules of grammar. Moderately to highly frequent argument NP orderings are likely have attained this status, but not their zero-frequency counterparts. This is why the latter argument sequences cannot be produced by the grammatical encoder and are absent from the corpora. Secondly, we assumed that an extraneous (nonlinguistic) judgment process biases the ratings of moderately grammatical linear order patterns: Confronted with such structures, the informants produce their own “ideal delivery” variant of the to-be-rated target sentence and evaluate the similarity between the two versions. A high similarity score yielded by this judgment then exerts a positive bias on the grammaticality rating—a score that should not be mistaken for an authentic grammaticality rating. We conclude that, at least in the linearization domain studied here, the goal of gaining a clear view of the internal grammar of language users is best served by a combined strategy in which grammar rules are founded on structures that elicit moderate to high grammaticality ratings and attain at least moderate usage frequencies.

Abstract

Four probe latency experiments show that the ‘constituent boundary effect’ (transitions between constituents are more difficult than within constituents) is a retrieval and not a storage phenomenon. The experimental logic used is called paraphrastic reproduction: after verbatim memorization of some sentences, subjects were instructed to reproduce them both in their original wording and in the form of sentences that, whilst preserving the original meaning, embodied different syntactic constructions. Syntactic constructions are defined as pairs which consist of a pattern of conceptual information and a syntactic scheme, i.e. a sequence of syntactic word categories and function words. For example, the sequence noun + finite intransitive main verb (‘John runs’) expresses a conceptual actor-action relationship. It is proposed that for each overlearned and simple syntactic construction there exists a retrieval plan which does the following. It searches through the long-term memory information that has been designated as the conceptual content of the utterance(s) to be produced, looking for a token of its conceptual pattern. The retrieved information is then cast into the format of its syntactic scheme. The organization of such plans is held responsible for the constituent boundary effect.

Abstract

The ”download first, then process paradigm” is still the predominant working method amongst the research community. The web-based paradigm, however, offers many advantages from a tool development and data management perspective as they allow a quick adaptation to changing research environments. Moreover, new ways of combining tools and data are increasingly becoming available and will eventually enable a true web-based workflow approach, thus challenging the ”download first, then process” paradigm. The necessary infrastructure for managing, exploring and enriching language resources via the Web will need to be delivered by projects like CLARIN and DARIAH

Abstract

In this paper, we describe a unifying approach to tackle data heterogeneity issues for lexica and related resources. We present LEXUS, our software that implements the Lexical Markup Framework (LMF) to uniformly describe and manage lexica of different structures. LEXUS also makes use of a central Data Category Registry (DCR) to address terminological issues with regard to linguistic concepts as well as the handling of working and object languages. Finally, we report on ViCoS, a LEXUS extension, providing support for the definition of arbitrary semantic relations between lexical entries or parts thereof.

Abstract

To achieve true interoperability for valuable linguistic resources different levels of variation need to be addressed. ISO Technical Committee 37, Terminology and other language and content resources, is developing a Data Category Registry. This registry will provide a reusable set of data categories. A new implementation, dubbed ISOcat, of the registry is currently under construction. This paper shortly describes the new data model for data categories that will be introduced in this implementation. It goes on with a sketch of the standardization process. Completed data categories can be reused by the community. This is done by either making a selection of data categories using the ISOcat web interface, or by other tools which interact with the ISOcat system using one of its various Application Programming Interfaces. Linguistic resources that use data categories from the registry should include persistent references, e.g. in the metadata or schemata of the resource, which point back to their origin. These data category references can then be used to determine if two or more resources share common semantics, thus providing a level of interoperability close to the source data and a promising layer for semantic alignment on higher levels

Abstract

Two Event-Related Potential (ERP) studies contrast the processing of locally ambiguous sentences in the visual and the auditory modality. These sentences are disambiguated by a lexical element. Before this element appears in a sentence, the sentence can also be disambiguated by a boundary marker: a comma in the visual modality, or a prosodic break in the auditory modality. Previous studies have shown that a specific ERP component, the Closure Positive Shift (CPS), can be elicited by these markers. The results of the present studies show that both the comma and the prosodic break disambiguate the ambiguous sentences before the critical lexical element, despite the fact that a clear CPS is only found in the auditory modality. Comma and prosodic break thus have parallel functions irrespective of whether they do or do not elicit a CPS.

Abstract

Functional imaging studies have demonstrated involvement of the anterior temporal cortex in sentence comprehension. It is unclear, however, whether the anterior temporal cortex is essential for this function.We studied two aspects of sentence comprehension, namely syntactic and prosodic comprehension in temporal lobe epilepsy patients who were candidates for resection of the anterior temporal lobe. Methods: Temporal lobe epilepsy patients (n = 32) with normal (left) language dominance were tested on syntactic and prosodic comprehension before and after removal of the anterior temporal cortex. The prosodic comprehension test was also compared with performance of healthy control subjects (n = 47) before surgery. Results: Overall, temporal lobe epilepsy patients did not differ from healthy controls in syntactic and prosodic comprehension before surgery. They did perform less well on an affective prosody task. Post-operative testing revealed that syntactic and prosodic comprehension did not change after removal of the anterior temporal cortex. Discussion: The unchanged performance on syntactic and prosodic comprehension after removal of the anterior temporal cortex suggests that this area is not indispensable for sentence comprehension functions in temporal epilepsy patients. Potential implications for the postulated role of the anterior temporal lobe in the healthy brain are discussed.

Abstract

The present article reports on an investigation of one child's acquisition of the multiple senses of the preposition with from 2;0–4;0. Two competing claims regarding children's early representation and subsequent acquisition of with were investigated. The “multiple meanings” hypothesis predicts that children form individual form-meaning pairings for with as separate lexical entries. The “monosemy approach” (McKercher 2001) claims that children apply a unitary meaning by abstracting core features early in acquisition. The child's (“Brian”) speech and his input were coded according to eight distinguishable senses of with. The results showed that Brian first acquired the senses that were most frequent in the input (accompaniment, attribute, and instrument). Less common senses took much longer to emerge. A detailed analysis of the input showed that a variety of clues are available that potentially enable the child to distinguish among high frequency senses. The acquisition data suggested that the child initially applied a restricted one-to-one form-meaning mapping for with, which is argued to reflect the spatial properties of the preposition. On the basis of these results it is argued that neither the monosemy nor the multiple meanings approach can fully explain the data, but that the results are best explained by a combination of word learning principles and children's ability to categorize the contextual properties of each sense's use in the ambient language.

Abstract

Hartshorne and Ullman (2006) presented naturalistic language data from 25 children (15 boys, 10 girls) and showed that girls produced more past tense overregularization errors than did boys. In particular, girls were more likely to overregularize irregular verbs whose stems share phonological similarities with regular verbs. It was argued that the result supported the Declarative/Procedural model of language, a neuropsychological analogue of the dual-route approach to language. In the current study we present experimental data that are inconsistent with these naturalistic data. Eighty children (40 males, 40 females) aged 5;0–6;9 completed a past tense elicitation task, a test of declarative memory, and a test of non-verbal intelligence. The results revealed no sex differences on any of the measures. Instead, the best predictors of overregularization rates were item-level features of the test verbs. We discuss the results within the context of dual versus single route debate on past tense acquisition

Abstract

To segment continuous speech into its component words, listeners make use of language rhythm; because rhythm differs across languages, so do the segmentation procedures which listeners use. For each of stress-, syllable-and mora-based rhythmic structure, perceptual experiments have led to the discovery of corresponding segmentation procedures. In the case of mora-based rhythm, similar segmentation has been demonstrated in the otherwise unrelated languages Japanese and Telugu; segmentation based on syllable rhythm, however, has been previously demonstrated only for European languages from the Romance family. We here report two target detection experiments in which Korean listeners, presented with speech in Korean and in French, displayed patterns of segmentation like those previously observed in analogous experiments with French listeners. The Korean listeners' accuracy in detecting word-initial target fragments in either language was significantly higher when the fragments corresponded exactly to a syllable in the input than when the fragments were smaller or larger than a syllable. We conclude that Korean and French listeners can call on similar procedures for segmenting speech, and we further propose that perceptual tests of speech segmentation provide a valuable accompaniment to acoustic analyses for establishing languages' rhythmic class membership.

Abstract

There are two core phenomena which any empirical investigation of beauty must account for: the existence of aesthetical experience, and the enormous variability of this experience across times, cultures, people. Hence, it would seem a hopeless enterprise to determine ‘the very nature’ of beauty, and in fact, none of the many attempts from the Antiquity to present days found general acceptance. But what we should be able to investigate and understand is how properties of people, for example their varying cultural experiences, are correlated with the properties of objects which we evaluate. Beauty is neither only in the eye of the observer nor only in the objects which it sees - it is in the way in which specific observers see specific objects.

Abstract

All disciplines depend on language; but two of them also have language as an object – literary studies and linguistics. Their objectives are not the same – but they are sufficiently similar to invite close cooperation. This is not what we find; in fact, the development of research over the last decades has led to a relationship which is, in the typical case, characterised by friendly, and sometimes less friendly, ignorance and indifference. This article discusses some of the reasons for this development, and it suggests some conditions under which both sides would benefit from more cooperation.

Abstract

This paper explores the relationship between the acoustic duration of phonemic sequences and their frequencies of occurrence. The data were obtained from large (sub)corpora of spontaneous speech in Dutch, English, German, and Italian. Acoustic duration of an n-phone is shown to codetermine the n-phone's frequency of use, such that languages preferentially use diphones and triphones that are neither very long nor very short. The observed distributions are well approximated by a theoretical function that quantifies the concurrent action of the self-regulatory processes of minimization of articulatory effort and minimization of perception effort

Abstract

How humans understand the intention of others’ actions remains controversial. Some authors have suggested that intentions are recognized by means of a motor simulation of the observed action with the mirror-neuron system [1–3]. Others emphasize that intention recognition is an inferential process, often called ‘‘mentalizing’’ or employing a ‘‘theory of mind,’’ which activates areas well outside the motor system [4–6]. Here, we assessed the contribution of brain regions involved in motor simulation and mentalizing for understanding action intentions via functional brain imaging. Results show that the inferior frontal gyrus (part of the mirror-neuron system) processes the intentionality of an observed action on the basis of the visual properties of the action, irrespective of whether the subject paid attention to the intention or not. Conversely, brain areas that are part of a ‘‘mentalizing’’ network become active when subjects reflect about the intentionality of an observed action, but they are largely insensitive to the visual properties of the observed action. This supports the hypothesis that motor simulation and mentalizing have distinct but complementary functions for the recognition of others’ intentions.

Abstract

Chronic fatigue syndrome (CFS) is a disabling disorder, characterized by persistent or relapsing fatigue. Recent studies have detected a decrease in cortical grey matter volume in patients with CFS, but it is unclear whether this cerebral atrophy constitutes a cause or a consequence of the disease. Cognitive behavioural therapy (CBT) is an effective behavioural intervention for CFS, which combines a rehabilitative approach of a graded increase in physical activity with a psychological approach that addresses thoughts and beliefs about CFS which may impair recovery. Here, we test the hypothesis that cerebral atrophy may be a reversible state that can ameliorate with successful CBT. We have quantified cerebral structural changes in 22 CFS patients that underwent CBT and 22 healthy control participants. At baseline, CFS patients had significantly lower grey matter volume than healthy control participants. CBT intervention led to a significant improvement in health status, physical activity and cognitive performance. Crucially, CFS patients showed a significant increase in grey matter volume, localized in the lateral prefrontal cortex. This change in cerebral volume was related to improvements in cognitive speed in the CFS patients. Our findings indicate that the cerebral atrophy associated with CFS is partially reversed after effective CBT. This result provides an example of macroscopic cortical plasticity in the adult human brain, demonstrating a surprisingly dynamic relation between behavioural state and cerebral anatomy. Furthermore, our results reveal a possible neurobiological substrate of psychotherapeutic treatment.

Abstract

Kanazawa (2007) proposes the ‘evolutionary psychological imagination’ (p.7) as an authoritative framework for understanding complex social and public issues. As a case study of this approach, Kanazawa addresses acts of international terrorism, specifically suicide bombings committed by Muslim men. It is proposed that a comprehensive explanation of such acts can be gained from taking an evolutionary perspective armed with only three points of cultural knowledge: 1. Muslims are exceptionally polygynous, 2. Muslim men believe they will gain reproductive access to 72 virgins if they die as a martyr and 3. Muslim men have limited access to pornography, which might otherwise relieve the tension built up from intra-sexual competition. We agree with Kanazawa that evolutionary models of human behaviour can contribute to our understanding of even the most complex social issues. However, Kanazawa’s case study, of what he refers to as ‘World War III’, rests on a flawed theoretical argument, lacks empirical backing, and holds little in the way of explanatory power.

Abstract

Digital processing of medical images has helped physicians and patients during past years by allowing examination and diagnosis on a very precise level. Nowadays possibly the biggest deal of support it can offer for modern healthcare is the use of high performance computing architectures to treat the huge amounts of data that can be collected by modern acquisition devices. This paper presents a parallel processing implementation of an image segmentation algorithm that operates on a computer cluster equipped with 10 processing units. Thanks to well-organized distribution of the workload we manage to significantly shorten the execution time of the developed algorithm and reach a performance gain very close to linear.

Abstract

Contains: Vol. 1 An introduction to the theory of formal languages and automata Vol. 2 Applications in linguistic theory Vol. 3 Psycholinguistic applications

Abstract

The present paper studies how, in deictic expressions, the temporal interdependency of speech and gesture is realized in the course of motor planning and execution. Two theoretical positions were compared. On the “interactive” view the temporal parameters of speech and gesture are claimed to be the result of feedback between the two systems throughout the phases of motor planning and execution. The alternative “ballistic” view, however, predicts that the two systems are independent during the phase of motor execution, the temporal parameters having been preestablished in the planning phase. In four experiments subjects were requested to indicate which of an array of referent lights was momentarily illuminated. This was done by pointing to the light and/or by using a deictic expression (this/that light). The temporal and spatial course of the pointing movement was automatically registered by means of a Selspot opto-electronic system. By analyzing the moments of gesture initiation and apex, and relating them to the moments of speech onset, it was possible to show that, for deictic expressions, the ballistic view is very nearly correct.

Abstract

This paper describes the descriptive landscape and seascape terminology of an isolate language, Yélî Dnye, spoken on a remote island off Papua New Guinea. The terminology reveals an ontology of landscape terms fundamentally mismatching that in European languages, and in current GIS applications. These landscape terms, and a rich set of seascape terms, provide the ontological basis for toponyms across subdomains. Considering what motivates landscape categorization, three factors are considered: perceptual salience, human affordance and use, and cultural ideas. The data show that cultural ideas and practices are the major categorizing force: they directly impact the ecology with environmental artifacts, construct religious ideas which play a major role in the use of the environment and its naming, and provide abstract cultural templates which organize large portions of vocabulary across subdomains.

Abstract

Comment on Sperber and Wilson

Abstract

Chinese translation of the 2003 publication.

Abstract

This paper is one in a series that develops a pragmatic framework in loose confederation with Jay Atlas and Larry Horn: thus they may or may not be responsible for the ideas contained herein. Jay Atlas provided many comments which I have utilized or perverted as the case may be. The Australian data to which this framework is applied was collected with the financial and personal assistance of many people and agencies acknowledged separately below; but I must single out for special thanks John Haviland, who recommended the study of Guugu Yimidhirr anaphora to me and upon whose grammatical work on Guugu Yimidhirr this paper is but a minor (and perhaps flawed) elaboration. A grant from the British Academy allowed me to visit Haviland in September 1986 to discuss many aspects of Guugu Yimidhirr with him, and I am most grateful to the Academy for funding this trip and to Haviland for generously making available his time, his texts (from which I have drawn many examples, not always with specific acknowledgement) and most especially his expertise. Where I have diverged from his opinion I may well learn to regret it. I must also thank Nigel Vincent for putting me in touch with a number of recent relevant developments in syntax (only some of which I have been able to address) and for suggestions for numerous improvements. In addition, I have benefited immensely for comments on a distinct but related paper (Levinson, 1987) kindly provided by Jay Atlas, John Haviland, John Heritage, Phil Johnson-Laird, John Lyons, Tanya Reinhart, Emanuel Schegloff and an anonymous referee; and from comments on this paper by participants in the Cambridge Linguistics Department seminar where it was first presented (especial thanks to John Lyons and Huang Yan for further comments, and Mary Smith for a counter-example). Despite all this help, there are sure to be errors of data and analysis that I have persisted in. Aid in gathering the Australian data is acknowledged separately below.

Abstract

This entry contains: 1. An invitation to think about to what extent the grammar of space and time share lexical and morphosyntactic resources − the suggestions here are only prompts, since it would take a long questionnaire to fully explore this; 2. A suggestion about how to collect gestural data that might show us to what extent the spatial and temporal domains, have a psychological continuity. This is really the goal − but you need to do the linguistic work first or in addition. The goal of this task is to explore the extent to which time is conceptualised on a spatial basis.

Abstract

In the current study we investigated whether 12-month-old infants gesture appropriately for knowledgeable versus ignorant partners, in order to provide them with needed information. In two experiments we found that in response to a searching adult, 12-month-olds pointed more often to an object whose location the adult did not know and thus needed information to find (she had not seen it fall down just previously) than to an object whose location she knew and thus did not need information to find (she had watched it fall down just previously). These results demonstrate that, in contrast to classic views of infant communication, infants’ early pointing at 12 months is already premised on an understanding of others’ knowledge and ignorance, along with a prosocial motive to help others by providing needed information.

Abstract

This paper investigates the social-cognitive and motivational complexities underlying prelinguistic infants' gestural communication. With regard to deictic referential gestures, new and recent experimental evidence shows that infant pointing is a complex communicative act based on social-cognitive skills and cooperative motives. With regard to infant representational gestures, findings suggest the need to re-interpret these gestures as initially non-symbolic gestural social acts. Based on the available empirical evidence, the paper argues that deictic referential communication emerges as a foundation of human communication first in gestures, already before language. Representational symbolic communication, instead, emerges as a transformation of deictic communication first in the vocal modality and, perhaps, in gestures through non-symbolic, socially situated routines.

Abstract

Young children demonstrate the ability to make inferences about the preferences of other agents based on their choices. However, there exists no overarching account of what children are doing when they learn about preferences or how they use that knowledge. We use a rational model of preference learning, drawing on ideas from economics and computer science, to explain the behavior of children in several recent experiments. Specifically, we show how a simple econometric model can be extended to capture two- to four-year-olds’ use of statistical information in inferring preferences, and their generalization of these preferences.

Abstract

We examined how communicators can switch between speaker and listener role with such accurate timing. During conversations, the majority of role transitions happens with a gap or overlap of only a few hundred milliseconds. This suggests that listeners can predict when the turn of the current speaker is going to end. Our hypothesis is that listeners know when a turn ends because they know how it ends. Anticipating the last words of a turn can help the next speaker in predicting when the turn will end, and also in anticipating the content of the turn, so that an appropriate response can be prepared in advance. We used the stimuli material of an earlier experiment (De Ruiter, Mitterer & Enfield, 2006), in which subjects were listening to turns from natural conversations and had to press a button exactly when the turn they were listening to ended. In the present experiment, we investigated if the subjects can complete those turns when only an initial fragment of the turn is presented to them. We found that the subjects made better predictions about the last words of those turns that had more accurate responses in the earlier button press experiment.

Abstract

This Field Manual entry has been superceded by the 2009 version:
https://doi.org/10.17617/2.883570

Abstract

The cross-linguistic investigation of semantic categories has a long history, spanning many disciplines and covering many domains. But the extent to which semantic categories are universal or language-specific remains highly controversial. Focusing on the domain of events involving material destruction (“cutting and breaking” events, for short), this study investigates how speakers of different languages implicitly categorize such events through the verbs they use to talk about them. Speakers of 28 typologically, genetically and geographically diverse languages were asked to describe the events shown in a set of videoclips, and the distribution of their verbs across the events was analyzed with multivariate statistics. The results show that there is considerable agreement across languages in the dimensions along which cutting and breaking events are distinguished, although there is variation in the number of categories and the placement of their boundaries. This suggests that there are strong constraints in human event categorization, and that variation is played out within a restricted semantic space.

Abstract

Coherent covariation appears to be a powerful explanatory factor accounting for a range of phenomena in semantic cognition. But its role in accounting for the crosslinguistic facts is less clear. Variation in naming, within the same semantic domain, raises vexing questions about the necessary parameters needed to account for the basic facts underlying categorization.

Abstract

Recurrent lexicalization patterns across widely different cultural contexts can provide a window onto common conceptualizations. The cross-linguistic data support the idea that sweet, salt, sour, and bitter are basic tastes. In addition, umami and fatty are likely basic tastes, as well.

Abstract

In this task we aim to find what the best exemplars or “focal colours” of each basic colour term is in our field languages. This is an important part of the evidence we need in order to understand the colour data collected using 'The Language of Vision I: Colour'. This task consists of an experiment where participants pick out the best exemplar for the colour terms in their language. The goal is to establish language specific focal colours.

Abstract

We present a computational model that provides a unified account of inference, coherence, and disambiguation. It simulates how the build-up of coherence in text leads to the knowledge-based resolution of referential ambiguity. Possible interpretations of an ambiguity are represented by centers of gravity in a high-dimensional space. The unresolved ambiguity forms a vector in the same space. This vector is attracted by the centers of gravity, while also being affected by context information and world knowledge. When the vector reaches one of the centers of gravity, the ambiguity is resolved to the corresponding interpretation. The model accounts for reading time and error rate data from experiments on ambiguous pronoun resolution and explains the effects of context informativeness, anaphor type, and processing depth. It shows how implicit causality can have an early effect during reading. A novel prediction is that ambiguities can remain unresolved if there is insufficient disambiguating information.

Abstract

What drives humans around the world to converge in certain ways in their naming while diverging dramatically in others? We studied how naming patterns are constrained by investigating whether labeling of human locomotion reflects the biomechanical discontinuity between walking and running gaits. Similarity judgments of a student locomoting on a treadmill at different slopes and speeds revealed perception of this discontinuity. Naming judgments of the same clips by speakers of English, Japanese, Spanish, and Dutch showed lexical distinctions between walking and running consistent with the perceived discontinuity. Typicality judgments showed that major gait terms of the four languages share goodness-of-example gradients. These data demonstrate that naming reflects the biomechanical discontinuity between walking and running and that shared elements of naming can arise from correlations among stimulus properties that are dynamic and fleeting. The results support the proposal that converging naming patterns reflect structure in the world, not only acts of construction by observers.

Abstract

Interpreting a verb-phrase ellipsis (VP ellipsis) requires accessing an antecedent in memory, and then integrating a representation of this antecedent into the local context. We investigated the online interpretation of VP ellipsis in an eye-tracking experiment and four speed–accuracy tradeoff experiments. To investigate whether the antecedent for a VP ellipsis is accessed with a search or direct-access retrieval process, Experiments 1 and 2 measured the effect of the distance between an ellipsis and its antecedent on the speed and accuracy of comprehension. Accuracy was lower with longer distances, indicating that interpolated material reduced the quality of retrieved information about the antecedent. However, contra a search process, distance did not affect the speed of interpreting ellipsis. This pattern suggests that antecedent representations are content-addressable and retrieved with a direct-access process. To determine whether interpreting ellipsis involves copying antecedent information into the ellipsis site, Experiments 3–5 manipulated the length and complexity of the antecedent. Some types of antecedent complexity lowered accuracy, notably, the number of discourse entities in the antecedent. However, neither antecedent length nor complexity affected the speed of interpreting the ellipsis. This pattern is inconsistent with a copy operation, and it suggests that ellipsis interpretation may involve a pointer to extant structures in memory.