Raul Aranovich

University of California, Davis | UCD · Department of Linguistics

An OWL ontology is used to model a grammar that accounts for subcategorization, showing that ontologies are able to generate (mildly) context-sensitive languages. Semantic Web knowledge representation methods offer a useful way to model the implicit knowledge that defines human linguistic abilities. When a grammar is modeled as a set of ontological...

Ferdinand de Saussure’s Cours de linguistique générale reproduced a misconception of Chinese as a monosyllabic language without complex words. In this paper, we investigate the sources of this misconception in Western thought. We also show that the misconception about Chinese was already known to be inaccurate in Saussure’s time, and that he had ma...

Cybersecurity affects us all in our daily lives. New knowledge on best practices, new vulnerabilities, and timely fixes for cybersecurity issues is growing super-linearly, and is spread across numerous, heterogeneous sources. Because of that, community contribution-based, question and answer sites have become clearinghouses for cybersecurity-relate...

Cybersecurity experts rely on the knowledge stored in databases like the NVD to do their work, but these are not the only sources of information about threats and vulnerabilities. Much of that information flows through social media channels. In this paper we argue that security experts and general users alike can benefit from the technologies of th...

Although the Neural Machine Translation (NMT) framework has already been shown effective in large training data scenarios, it is less effective for low-resource conditions. To improve NMT performance in a low-resource setting, we extend the high-quality training data by generating a pseudo bilingual dataset and then filtering out low-quality alignm...

This paper describes a systematic study of an approach to Farsi-Spanish low-resource Neural Machine Translation (NMT) that leverages monolingual data for joint learning of forward and backward translation models. As is standard for NMT systems, the training process begins using two pre-trained translation models that are iteratively updated by decr...

Neural Machine Translation (NMT) performs training of a neural network employing an encoder-decoder architecture. However, the quality of the neural-based translations predominantly depends on the availability of a large amount of bilingual training dataset. In this paper, we explore the performance of translations predicted by attention-based NMT...

In this paper, we propose a useful optimization method for low-resource Neural Machine Translation (NMT) by investigating the effectiveness of multiple neural network optimization algorithms. Our results confirm that applying the proposed optimization method on English-Persian translation can exceed translation quality compared to the English-Persi...

In the well-known analysis of Cinque (1988), the generalization that a language will form impersonal reflexives of unaccusatives if and only if it can have accusative pivots follows from Parameter Theory. But the Balkans provide a two-way exception to this prediction. First, impersonal reflexives of unaccusatives are found in Standard Bosnian, Croa...

In Spanish ditransitive clauses with Dative Clitic Doubling (DClD), the goal may be doubled by a dative clitic pronoun. Applying a chi-square decomposition analysis to examples with the prototypical ditransitive verb dar “give”, from a corpus of journalistic language characteristic of River Plate Spanish, I show that DClD is disfavoured when the th...

Poster given at the 2016 LSA conference on Shona periphrastic verbs with the auxiliary -ti ’do’.

Poster given at the 2016 LSA conference on Shona periphrastic verbs with the auxiliary -ti ’do’.

Poster given at the 2016 LSA conference on Shona periphrastic verbs with the auxiliary -ti ’do’.

In this paper, I argue that the choice of a periphrastic over a synthetic verbal form in Shona is a strategy to avoid the realization of specific combinations of morphosyntactic properties by the same word form. I argue that periphrastic forms with -nga are formally equivalent to synthetic forms in the verbal paradigm of Shona, based on the assumpt...

In Item-and-Arrangement models of inflection, morphemes are associations of form and meaning stored in a mental lexicon. Saussure’s notion of the linguistic sign as a unit of an acoustic image (signifier) and a concept (signified) immediately suggests such a model. But close examination of the examples of inflectional morphology throughout the Cour...

Proceedings of the Twenty-First Annual Meeting of the Berkeley Linguistics Society: General Session and Parasession on Historical Issues in Sociolinguistics/Social Issues in Historical Linguistics (1995)

This paper argues for a three-way structural characterization of Fijian objects: common nouns can be incorporated or dislocated, but pronouns and proper nouns occur inside the VP as complements. These facts support an analysis of Fijian as a polysynthetic language, since it is a pronominal argument language with incorporated objects. Having complem...

This paper explores the interaction between Dative Clitic Doubling (DClD) in Spanish ditransitives and some semantic and syntactic factors. Usage data give statistically significant evidence that a goal is more likely to be doubled when definite or when it precedes the theme. These facts are accounted for under the hypothesis that goals alternate b...

Across the Romance languages, impersonal reflexives vary with respect to which objects can control agreement. This paper offers an account of such variation within Optimality Theory. In languages that exclude accusative-marked objects from impersonal reflexives, a constraint favoring agreement outranks those constraints responsible for differential...

Languages have differential subject marking (DSM) when case marking, agreement, or constituent order codes some subjects differently from others. Work on split ergative languages provides much of what is already known about DSM, as does research on dative or genitive subjects (see Aikhenvald et al. 2001 for a recent collection of papers on this top...

Analysis of the alternation between impersonal reflexives and reflexive passives in Spanish shows that higer animacy makes an argument less marked as an object. This generalization supports Hopper and Thompson's (1981) Transitivity Hypothesis, since it is entailed by their claim that objects tend to be highly individuated. The paper proposes an Opt...

Les parfaits des intransitifs avec ÊTRE dans les langues romanes tirent leur origine des parfaits des verbes déponents latins. Cependant, dans certains cas, il arrive qu'un verbe déponent latin évolue en un intransitif sélectionnant AVOIR, ou encore qu'un parfait roman utilisant ÊTRE ne soit équivalent à aucun déponent latin. Afin d'expliquer de te...

In the Mordvin definite conjugation, portmanteau verbal affixes cross-reference the person and number of both subject and object. This complex agreement system shows extensive syncretism, which is more acute among the most marked members of the paradigm. I argue that Optimality-Theory gives a more explanatory account of the pattern of neutralizatio...

Licensing of NPIs in Spanish varies depending on the semantics of the trigger. Nonveridical operators license n-words, and antiveridical operators license ni-minimizers. I argue that the NPIs that can occur in antiveridical contexts have a scalar presupposition, but those that are licensed in strictly nonveridical contexts are non-scalar. In this a...

Wasow (1977) argues that linguistic theory should recognize two qualitatively distinct types of rules: syntactic rules, which can affect more "superficial" grammatical function properties; and lexical rules, which affect deeper lexical semantic properties of lexical items. However, lexicalist theories of grammar have replaced syntactic rules with l...

Old Spanish had a split auxiliary system in the perfect tense, reminiscent of what is found in Modern French and Modern Italian. In this paper, I trace the progress of the displacement of ser ‘be’ by haber ‘have’ with intransitive and reflexive verbs in the history of Spanish. The data support the hypothesis that predicates that have a more patient...

This paper argues for a semantic approach to split intransitivity based on the distribution of the embedded infinitives in the quedar-por construction in Spanish. Typical unaccusatives can appear in this construction, but it is noticed that only a sub-class of reflexive verbs can do so. The paper shows this result to be unexpected for a syntactic a...

Thesis (Ph. D.)--University of California, San Diego, 1996. Vita. Includes bibliographical references (leaves 308-317).

1.SG.PAST-leave-APPL-F child present 'I left the gift for the child.' 'I left the gift for the child.' In symmetrical languages, the Asymmetrical Object Parameter (AOP) is set to allow for two internal arguments to be intrinsically classified as (-r) (unrestricted). The analysis of Shona as a symmetric language is challenged by an animacy constrain...