Linguistics Free Word Order, Scrambling

Linguistics: Free Word Order, Scrambling

This work conducts a review of historical and recent literature related to ‘free word order’ languages, or those, which use ‘scrambling’ in sentence structure placement of nouns and verbs. The social theory of language acquisition is reviewed as well as cultural influences on language acquisition and specifically related to ‘free word order’ language structure. Word order in various languages is examined and recent studies reviewed in this subject area.

Linguistics: Free Word Order, Scrambling

The objective of this work is to review both historical and recent literature related to ‘free word order’ or ‘scrambling’ in languages such as Japanese and Korean in relation to the effect that this scrambled order of nouns and verbs in the sentence structure has upon the learning of the language by those native to these languages. This work will further research and review the various theories associated with language acquisition in free word order structured languages. This work does not attempt to provide an explanation but only to review the recent work in this subject area

Introduction

Free word order’ languages are those in which the structure of a sentence is constructed loosely in relation to the placement of nouns and verbs. The writer notes in the research process that there are many and various explanations and theories surrounding the structure or lack of structure in the free word order languages. Grammatical encoding has never been quite as relevant as in the present as computer-generated language translation is in use frequently in the lives of many. Communication barriers presented are evident in the confused communications and specifically between languages such as the English language with its formally structured sentence use of nouns and verbs and the languages of Korean and Japanese both ‘free word order’ languages..

Thorne (2000) notes in his work “Second Language Acquisition Theory and the Truths About Relativity” that strong tradition in research exists within anthropology, sociology and linguistics which illustrate the contextual relativity of semantic values. Reappraising these values in relation to second language acquisition from “an understanding of the language learned as context-independent lexical and grammatical meaning (present in formal theories of language and communication), to an acknowledgment of the relative and context contingent nature of language-in-use.” (Thorne, 2000)

Thorne remarks that when socio-cultural theory is afforded a serious view from the perspective of: “…linguistics, anthropology, sociology, psychology, and communications theory as they pertain to second and foreign language learning” that certain themes unfold. These themes are those as follows:

1) the interdependences between language and conceptual development, 2) Language as the principal sign system carrying socio-historical-cultural presence into the moment, 3) Language as a primary resource through which people interactively construct social reality, and (4) the reproduction of individual and community practices due in part to the inertia of language constructed social worlds.” (Thorne, 2000)

I. Socio-Cultural Formation of Speech and Language

Vygotsky believed that the human brain is: “…socially constituted through mediation via semiotic systems” meaning specifically that the different languages are actually…expressions of socio-historical processes” existing in the life of the individual. To Vygotsky speech is first ‘social’ and then follows internalized or ‘inner’ speech referred to as ‘egocentric’ speech. Not only do considerable differences exist between the various languages throughout the world but also existing are “language specific world-framing” differences. (Thorne, 2000; paraphrased) in the attempt to learn a language adults specifically must learn in relation to “reduced syntax and vocabulary, no fixed order of words, with considerable variation from on e speaker to another; and used purely as a language of communication.” (Mason, nd)

It was held by Noam Chomsky that children are born with an innate knowledge or set of rules concerning language which he called “Universal Grammar’. Chomsky says that: Children are exposed to very little correctly formed language. When people speak, they constantly interrupt themselves; change their minds, makes slips of the tongue and so on. Yet children manage to learn their language all the same. This claim is usually referred to as the ‘Argument from Poverty of the Stimulus’. Children do not simply copy the language that they hear around them. They deduce rules from it, which they can then use to produce sentences that they have never heard before. They do not learn a repertoire of phrases and sayings, as the www2.carthage.edu/departments/teachba/” l “Table%20of%20Contents” behaviorists believe, but a grammar that generates infinity of new sentences. (Mason, nd) Bailyn (nd) states in the work entitled: “Word Order and Minimalism” that “Scrambling is the cover term of the transformation that derives non-canonical word order patterns in free word order languages such as Japanese, Russian, German, Hindi and others.” (Bailyn, nd) in these languages, constituents appear in a variety of surface orders, without changing the core meaning of sentence.” (Ibid)

II. Previous Study in Language Acquisition

Among the first to study acquisition of language were Gopnik and Choi (1991) followed by Gopnik and Meltzoff (1993). The studies of Gopnik and Choi (1991) demonstrate “specific correspondence between language-culture environment and conceptual development.” (Thorne, 2000) Thorne states that 75% of all languages in the world have a SVO structure, which includes English, French, and Vietnamese while others have a SOV structure, which includes Japanese and Korean. Other language structures are VOS and OSV. Children unconsciously recognize different language structures and make adjustments for them. (Thorne, 2000; paraphrased) it was demonstrated in the studies of Gopnik and Choi (1991) and Gopnik and Meltzoff (1993) that the Korean and Japanese language in terms of the language acquisition of children was greatly effected due to the culture-language environment and that specific correlations exist between grammatical characteristics of the languages and in relation to domain specific development progress.

III. Differentiation between ‘verb salient’ and ‘noun salient’ Languages

The Japanese and Korean languages are ‘verb salient’ languages while the English language is a ‘noun salient’ language. The Japanese and Korean children produced verb morphology earlier than did the children who spoke the English language however, the Korean and Japanese speaking children were “significantly delayed in performance concerning categorization tasks.” (Thorne, 2000) the work of Yama*****a and Chang entitled: “Sentence Production in Japanese” states that in order to verbally communicate with one another there is a requirement for the production of a: “…grammatical sequence of words, that is, a sentence that conveys a meaning. The meaning activates syntactic, morphological and lexical representations, which insure the production of a grammatical sentence.” (nd)

Yama*****a and Chang state that this process is one that is quite complex due to the fact that the individual speaking must: “select elements that are appropriate for the meaning and find a way to sequence these elements in a language-appropriate way.” (nd) in the initiative of providing an explanation of precisely how the individual who is speaking produces sentences then.”..an architecture for sentence production has been proposed (Bock, 1995; Bock & Levelt, 1994; Garrett 1998; Levelt, 1989) the three main points in this work in writing are:

1) the message – represents the “meaning of a sentence, 2) the grammatical component, responsible for the “generation of word sequences”; and 3) the physiological components which retrieves the content of worlds from the lexicon” (Chang and Yama*****a, 2006)

The grammatical characteristics of the Japanese language are reviewed in the work of Chang and Yama*****a (nd). The Japanese language is different from the English language in several “syntactic dimensions.” (Chang and Yama*****a, 2006) in fact, the Japanese language is one that utilizes ‘scrambling’ and is therefore “a relatively ‘free word-order’ language” the differences noted in the language of Japanese are those as follows:

1) Japanese language uses particles for marking each phrase in terms of its’ grammatical function within the sentence. This is an interchange of phrases within the sentence without changing the function of the phrase. The phrases trade places through scrambling or ‘word permutation’; (2) the verbs in the Japanese language are found at the clause ending; and (3) Arguments may be omitted within the framework of the Japanese language.” (Ibid)

IV. Word Order in the Various Languages

Keller writes that there is a substantial differentiation in the degree of variation of word order in the world’s languages as: “On one end of the spectrum, we find languages like English, which exhibit a relatively fixed word order. On the other end, there are languages like Warlpiri which allow a large degree of word order variation.” (Keller, nd) Keller further notes that “many languages exhibit a semi-free word order, ie.e the word order is fixed in some respects, but variable in others.” (nd) Weighted constraints were modeled by Uszkoreit (1987) who models word order preferences through annotation with numeric weights reflecting the importance toward grammaticality determination. Introduced the work of Keller is the “Optimality Theory” (Prince and Smolensky, 199) which assumes ” a binary notion of grammaticality, a linguistic structure is either optimal (and thus grammatical) or suboptimal (and thus ungrammatical) the work of Hawkins (1992) relates an approach to word order preference with a reliance, although under different assumptions in relation to the source of the competition stating that the constituent order is determined by the syntactic weight of the constituents.” (Keller, nd) Hawkins uses syntactic weight in explaining word order frequencies and the relative acceptability of different orders in native speakers’ judgments.” (Keller, nd)

The work of Christiansen (2002) entitled: “Case, Word Order, and Language Learnability: Insights from Connectionist Modeling” it is related that children learn “most of their native language within the first five years of life.” (2002) Christiansen further relates that the most difficult task in learning a language involves “mapping a sequence of words onto some sort of interpretation of what the sequence is supposed to mean.” (2002) in other words if the child is able to understand a sentenced then the child “needs to determine the grammatical roles of the individual words so that she can work out who did what to whom.”(Ibid) the work of Saffran, Aslin & Newport (199) acknowledges linguistic universals that are common “across radically different languages” and which indicate existing inherent learning constraints.

The two primary methods that signals within a language point to the syntactic relationships and grammatical roles are those of (1) word order (WO) and (2) case markings. English, which is strictly a word order language contains declarative sentences that are based on a Subject-Verb Object (SVO) patterns. In comparison, the Japanese language contains multiple word orders and depends on case markings to identify subjects from objects. Language acquisition theory has long held that an innate language acquisition device is to be credited however, Christensen states that “an alternative approach that is gaining ground is the adaptation of linguistic structures to the human brain rather than vice versa.” (Christiansen, 1994, Kirby, 1998)

The work of Gertner, Fisher, and Eisengart (2006) entitled: “Learning Words and Rules: Abstract Knowledge of Word Order in Early Sentence Comprehension” states that while children are known to acquire basic grammatical facts in relation to their native language the question exists of whether “this early syntactic knowledge involves knowledge of ‘words’ or ‘rules’? State is that: “According to lexical accounts of acquisition, abstract syntactic and semantic categories are not primitive to the language-acquisition system; thus early language comprehension and production are based on verb-specific knowledge.” (Ibid)

The study of Gertner, Fisher and Eisengart (2006) examines the “abstractness of young children’s knowledge of syntax through conducting tests as to whether children age 25 and 21 months are able to extend knowledge of the words order in English to new verbs. Four experiments were conducted in which children “used word order appropriately to interpret transitive sentences containing novel verbs. Findings state that while toddlers “have much to learn about their native languages, the represent language experience in terms of an abstract mental vocabulary.” (Ibid) it is held in this work that is the “abstract representations” that provide the children the opportunity to detect “patterns of a general nature within their native language”(Ibid) as well as providing the opportunity to learn the rules and words in the start of their language acquisition.

The work of Harriet S. Wetstone and Bernard Z. Friedlander entitled: “The Effect of Word Order on Young Children’s Responses to Simple Questions and Commands” states that the tendency in language acquisition in increasingly viewing the language learning of children as the “development of a system of communication rather than as the unfolding of a formal system of syntax.” (Bloom, 1970; Kaplan & Kaplan, 1970; Lewis & Freedle, 1972; Macnamara, 1972; as cited by Wetstone and Friedlander, 1973) Wetstone and Friedlander write that: “One of the most puzzling aspects of language learning…is children’s apparent ability to unravel the enormous complexity of language structure despite the garbled and fragmentary nature of the ordinary ongoing conversation that constitutes the corpus of their listening experience.” (Ibid) it has been held that children simply “bypass some of the confusion of language simply by ruling out what does not communicate to them; the listen selectively to what has meaning and is familiar (Shipley, Smith & Gleitman, 1969) in the study of Wetstone and Friedlander 20 children ages 2 and years old were asked simple questions and given simple commands in the “context of an at-home play situation.” The commands and questions were verbalized in variations of word order for evaluation of the effectiveness of communication in relation to word order and the ability of children to comprehend the meaning of the words. While the children who were ‘nonfluent’ provided appropriate responses to normal and scrambled sentences the ‘fluent’ children scored significantly, lower to the scrambled sentences than they scored on their responses to sentences that were scrambled. This indicates that the receptive language processing of nonfluent children has as its’ focus the familiar semantic elements rather than focusing on syntactic framework.

Cho, et al. (2002) states that children in Korea:

typically manifest higher comprehension rates on the ‘unmarked’ SOV sentences of their language than on the ‘scrambled’ OSV patterns. However, scant attention has been paid to the ordering preferences of children in relation to direct and indirect objects.” (Ibid)

In an experiment involving 40 children between the ages of 4 and 7 a strong preference for accusative-dative order. It is stated that despite evidence that the reverse order is more common in mother-to-child speech. In this work there are two considered hypotheses: (1) involving the relationship between word order and grammatical relations; and (2) involving the relationship between word order and the types of situations denoted by the sentences in question.” (Cho, et al., 2002)

Inconsistent structures in language “are harder to learn than consistent structures by computational systems, whether inconsistencies are at the syntactic level or at the lexical level, in terms of grapheme to phoneme correspondences, or semantic ambiguities.” (Monaghan, Gonitzke, Chater, nd) it is pointed out in the work of Monagham, Gnitzke, and Chater (nd) that it has been illustrated that the various frequencies of linguistic structures “have an impact on their ease of processing.. [making it] “extremely useful to have a representation of the relative frequencies of different structures in languages in order to make assertions about the ease of acquisition of inconsistencies that may occur within the language.” (Ibid) Monahan nd Gonitzke state that two primary influences on the simple recurrent network’s ability to learn sequential orders exist and the first of which is (1) predictability in word order; and (1) the impact of centre-embeddings in structures.”

Frank Keller (nd) in the work entitled: “Evaluating Competition-based Models of Word Order” states that three different models have been used in the proposition of explaining word order preferences. These three models have as their basis: (a) Weighted constraints; (b) Optimality Theory; and – Syntactic weight.” (Keller, nd) Grammatical competition is utilized in providing an explanation of the constraints to the world order while relying on “intuitive judgments or corpus studies…” (Ibid) Keller relates that there is a substantial differences in the degree of variation of word order across the broad spectrum of the differing world languages. The variation of manifested in the preference of world order, which is affected by factors that are: (1) Syntactic; (2) Pragmatic; and (3) Phonological. (Keller, nd)

The German language is a “verb-second language” that is a language described as one in which “all non-finite verb forms appear at the very end of the clause, so that finite verb in second position and the non-finite ones in final position together forms a sort of bracket around the main body of the clause.” (Buring, 1999) Buring cites cf Lenerz (1977) and Uszkoreit (1987) as well as Muller (1998) and state that: “It has been observed that various factors determine the acceptability of a given word order in a particular case, among them case, definiteness, animacy, and focus.” Homan writes that: “Languages such as Catalan, Czech, Finnish, German, Hindi, Hungarian, Japanese, Polish Russian and Turkish…” languages are some of the languages that have a word order that is “much freer…than English.” (nd) Homan states that upon attempting to translate English text into a language that is one of ‘free’ word order that a choice is presented “between many different word orders that are all syntactically grammatical but are not all felicitous of contextually appropriate.” (nd)

V. Non-Configurationality in Languages

The work of Beockx (nd) states that Hale (1983) listed three primary properties which have been “tied to the notion of non-configurationality” in languages which are those of: (1) “Free word order; (2) the use of syntactically discontinuous expressions; and (3) Extensive use of null anaphora to capture these properties” (nd) More recently Kitahara (1999) has proposed an “alternative analysis of scrambling within the minimalist program.” It is noted by Kitahara that “Case checking makes room for semantically vacuous overt movement…” And may lead to “what appear to be radical reconstruction effects since the element is interpreted upon case-checking and interpretation.” Even more recently the work of Niinuma (2000) has used a variety of tests to demonstrate that long-distance scrambling is focus-driven” and makes the claim that “scrambling affects focus structure.” This work builds upon Stejapanovi (1999) in relation to showing that “long-distance scrambling interacts with focus.” If scrambling is indeed driven by focus or discourse consideration then the claims of “Kitahara and Miyagawa” may be restated as follows: “long-distance scrambling is driven by a [+focus] feature that requires movement to some specifier position (phrasal displacement). Although this renders their analyses fully consistent with current theoretical assumptions, it now makes scrambling Beockx (nd)holds that:.”..scrambling is a well-behaved syntactic phenomenon.” Stated as well is that: Chomsky made the proposition that a “presumptive pronoun and its antecedent start off as a big-XP constituent. The derivation of the surface word order is from the sub-extraction of the antecedent, which is quite similar to the work of Sportich (1988) which analyzed the Floated Quantifiers and in the work of McCloskey (2000) in relation to a-bar movement. The work of Chomsky (2000:122) states that there are three specific ways that features may interact which are those of:

1) Features match;

2) the properties of features trigger ‘agree’; and 3) the properties of features trigger ‘move’. (Ibid)

VI. Computer-generated Processing of languages

The work of Rambow and Joshi (1995) entitled: “A Processing Model for Free Word Order Languages,” is a report from the Institute for Research in Cognitive Science states that: “German is a verb-nal language” and “like many verb-nal languages, such as Hindi, Japanese and Korean, it displays considerable word-order freedom.” (Rambow and Joshi, 1995) the work of Mazuka, Itoh and Kondo (Ibid) entitled: “Costs of Scrambling in Japanese Sentence Processing” published online the MIT CogNet Brain Sciences Connection states that in linguistic terms the existence of sentences that are ‘scrambled’ have been used in the argument for “a flat non-configurational structure for Japanese. Hale, 1980; Farmer, 1980). Alternatively, the facts that some grammatical phenomena, such as pronominalization, scope interpretation, and weak crossover are affected by scrambling have been used to argue for a configurational structure of Japanese (e.g., Saito, 1985, Hoji, 1986;1992; Saito and Hoji, 1983)” (Mazuka, Itoh and Kondo, 2004) This work reports the study of Mazuka, Itoh and Kondo in which “scrambled and non-scrambled sentences are studied in four experiments. Two of the studies are questionnaire studies with a total of five studies conducted in all. The results from the studies are stated to have shown that scrambled sentences indeed show an increased processing load associated with them. Japanese speakers rated scrambled sentences (2) and (5) more difficult and more misleading compared to nonscrambled counterpart (1) and (3) respectively. However, sentence (4), in which the scrambled NP had the relative clause and thus did not have a center-embedded structure was significantly easier than (5). Eye movement data and the self-paced reading time data showed in general an increased processing cost for scrambled sentences. Critically, the second argument in a scrambled sentence (John-Nom in (2)) had significantly longer reading time than the second argument of the unscrambled sentence (Majuka, Itoh and Kondo, 1995)

Reported in Keller (2000) are the results of a study of variation in word order in the German language, which invested “the interaction f syntactic and information structural constraints. The data evaluated a set of competition-based models or word order which included a) Uszkoreit’s (1987) weighted constraint; model, b) Muller’s (1999) optimality theoretic account, and Hawkins’s (1992) syntactic weight model.” (Keller, 2000)

Keller (2000) states that: while the experimental data are:.”..broadly compatible with models (a) and (b), indicating that a relativized (ranked or weighted) notion of linguistic constraints is essential for explaining word order preferences. Model -, however, was not well-supported by the data. While this model may be suitable for describing word order distributions in corpora, it does not seem to be directly applicable to contextualized acceptability judgments such as the ones reported in the present paper. On the other hand, we found that some of the individual linguistic assumptions made by Uszkoreit and Muller were not born out in our data. This highlights the fact that informal acceptability judgments are not sufficient to clarify the intricate preference patterns that emerge from the interaction of syntactic, pragmatic, and phonological constraints on word order. Experimentally collected judgments are necessary to obtain reliable, delicate data that can inform detailed models of word order preferences. The results of the present study have been replicated for a free word order language (Greek) and for spoken stimuli (Keller & Alexopoulou, 2000)” (Keller, 2000)

Paul Smolensky states that: “According to Optimality Theory (OT), possible human languages share a common set of universal constraints on well-formedness. These constraints are highly general, and hence conflict; thus some must be violated in optimal, i.e., grammatical, structures. The different surface patterns of the world’s languages emerge via different priority rankings of the fixed set of universal constraints: each ranking is a language-particular grammar, a means of resolving the inherent conflicts among the universal constraints.”

Parker (2006) in the work: “Word Order and the Formalized Information Structure in Japanese” attempts to account for the word order variations in the Japanese languages and states that: “Whereas English uses position to indicate subjects and objects (e.g. “The man ate a bear” and “A bear ate the man”), languages such as Japanese and Turkish use suffixes or inflections on nouns to show difference in grammatical roles.” (Parker, 2006) According to Parker, in languages such as these the world order: “…is more freely varied, and “scrambling” constituents from their canonical positions is much more common. The effects of scrambling are subtler than in a relatively “fixed” word order language like English. Earlier analyses of scrambling in Japanese vaguely described the effect of fronting a constituent as adding “mild emphasis.” In order to formalize the effects of scrambling in “free” word order languages, Beryl Hoffman developed a heuristic she called Multiset CCG (‘Combinatory Categorical Grammars’). She uses Multiset CCG to simultaneously parse for argument structure (i.e. verb-noun relations) and information structure, the ranking of sentence elements according to their status as information in discourse. Hoffman originally applied Multiset CCG only to Turkish but claimed it could also be applied to other languages, including Japanese. In order to formalize the effects of scrambling in “free” word order languages, Beryl Hoffman developed a heuristic she called Multiset CCG (‘Combinatory Categorical Grammars’). She uses Multiset CCG to simultaneously parse for argument structure (i.e. verb-noun relations) and information structure, the ranking of sentence elements according to their status as information in discourse. Hoffman originally applied Multiset CCG only to Turkish but claimed it could also be applied to other languages, including Japanese.” (Parker, 2006) Parker states that the project conducted makes examination of Hoffman’s claim through making adjustment to the Multiset CCG and then applying it to the Japanese language. Parker states that he: “…arranged a small set of relevant sentence data to test Multiset CCG, manually parsed them, and then compared the information structures produced by Multiset CCG to other analyses of those same Japanese sentence patterns, such as Kuno 1973. Interestingly, Hoffman’s information structures largely corresponded to the other analyses. In particular, parsing revealed a somewhat systematic distinction of the different uses of the topic-marking particle and the subject-marking particle as described by Kuno (1973). Multiset CCG also described post-verbal constituents as “ground,” which is strikingly similar to extant interpretations of post-verbal “afterthought” expressions in Japanese.” (Parker, 2006) Further Parker states that research revealed that: “…although Hoffman’s Multiset CCG includes restrictions on its parsing rules to forbid certain ungrammatical structures, these rules did not catch all the ungrammatical structures in Japanese, which has a word order that is less “free” than Turkish.” (Parker, 2006)

Reported recently is that Jong-Bok Kim and Jaehtyung Yang are in the process of developing computations grammar for Korean which has as its aim the development of an open-source grammar of Korean. The source stated that “the advent of the information era in this century has escalated the importance of processing linguistic information more precisely and correctly. Recent developments in artificial intelligence, information sciences and other high technology activities have made it possible to build feasible computational applications for language processing and understanding.” (Korean Resource Grammar, 2006) Applications include:

1. Message extraction systems;

2. Web-based search engines;

3. Machine translation; and 4. Dialogue understanding systems (Ibid)

Requirements exist for increasing accuracy and robustness of the grammar combined with sophisticated processing methods.” (Korean Resource Grammar, 2006) Stated objectives for this project is the creation of a “general purpose system for processing the Korean language that will support both research and practical application.” (Ibid) Goals are stated to include: (1) building a broad-coverage Korean grammar that can be used both to extract precise meanings from text input; and (2) to Generate well-formed text output. (Ibid) the project intends to develop a “computationally feasible Korean Resource Grammar and implement it into the Linguistic Knowledge Building system developed by the LinGO (Linguistic Grammar Online) Lab researchers at the Center for Study of Language and Information.” (Ibid) Stated as areas for further development are those of:

Refining the current Korean Resource Grammar (pro drop, case, relativization, light verb constructions, etc.)

Developing fine-grained semantics using MRS that can capture scope, event structures, message types, linking between syntax and semantics)

Incrementally increasing coverage of clause internal syntax in Korean (e.g., coordination, different types of long-distance dependencies, pro-drop, honorification, tense, aspect, coordination, etc.)

Incorporating the use of default entries for words unknown to the Korean HSPG lexicon

Testing with real-time corpora and expanding more coverage; and Linking the Korean Resource Grammar and the English Resource Grammar for applications such as machine translation ” (Korean Resource Grammar, 2006)

Summary

It is interesting to note that while theorists still disagree in many aspects in free word order language theory that the information systems has proceeded forward into implementation of Korean Resource Grammar. This is most likely due to the necessity that drives the creation of information system linguistic applications it is certain that these applications are much more needed than ever before and that necessity in addition to convenience are the factors that have tended to drive development and specifically has been witnessed to be the primary driver behind all information systems development in recent years. This is no different in the area of linguistic information systems development.

Bibliography

Keller, Frank Institute for Communicating and Collaborative systems, Division of Informatics, University of Edinburgh. UK. Gleitman, Lila R. And Joshi, Aravid K. eds, (2000) Proceedings of the 22nd Annual Conference of the Cognitive Science Society. Mahawah, NJ; Lawrence Erlbaum, 2000.

Yama*****a, H. And Chang, F. (nd) Sentence Production in Japanese. Rochester Institute of Technology – Anthropology. Online available at http://email.eva.mpg.de/~chang/papers/jprodhandbook.pdf.

Rambow, O. And Joshi, a.K. (1995) a Processing Model for Free Word Order Languages” Institute for Research in Cognitive Science, IRCS Technical Reports Series. University of Pennsylvania, 1995. Online available at http://repository.upenn.edu/cgi/viewcontent.cgi?article=1130&context=ircs_reports..

Parker, Brian (2006) Word Order and Formalized Information Structure in Japanese Ohio University. Online available at http://aolsearch.aol.com/aol/search?query=definition%3A+free+word+order%2C+scrambling&page=13&nt=null&userid=4077659711316363726&encquery=99c5f34c598f1bdefa46eb3f4e6f3a1535bc8a175d9d4420d62fb8527be753e9ae3463cecdde6d88&ie=UTF-8&invocationType=keyword_rollover&clickstreamid=-2316888368559440924

Mazuka, R. Itoh, K. And Kondo T. (2004) Costs of Scrambling in Japanese Sentence Processing. MIT CogNet – the Brain Sciences Connection. Online available at http://cognet.mit.edu/library/conferences/paper-paper-_id=1825.

Gertner, Y., Fisher, Cy., Eisengart, J. (2006) Learning Words and Rules: Abstract Knowledge of Word Order in Early Sentence Comprehension. Psychological Science. Vol. 17, No. 8 August 2006. Online available at http://www.ingentaconnect.com/content/bpl/psci/2006/00000017/00000008/art00009.

Buring, Daniel (1999) Let’s Phrase it! Focus, Word Order, and Prosodic Phrasing in German Double Object Constructions. 2 December. Cited in Muller, Gereon & Wolfgang Stemefeld (eds)(2000) Competition in Syntax. Mouton/deGryter (Studies in Generative Grammar)

Homan, Beryl

Translating into Free Word Order Language. Center for Cognitive Science. Online available at http://www.essex.ac.uk/linguistics/clmt/papers/mt/9604020.ps.

Lupyan, Gary (2002) Case, Word Order, and Language Learnability: Insights from Connectionist Modeling. Cornell University. http://cnl.psych.cornell.edu/papers/LandC-cogsci2002.pdf

Wetstone, J. And Friedlander (197) the Effect of Word Order on Young Children’s Reponses to Simple Questions and Commands. Child Development Journal Vol., No. Dec 1973.

The work of Cho, et al. (2002) entitled: “Word Order Preferences for Direct and Indirect Objects in Children Learning Korean. Cambridge University Press 21 Nov 2002.

Monaghan, Padraic; Gonitzke, Markus; and Chater, Nick

Two Wrongs Make a Right: Learnability and Word Order Consistency. Department of Psychology. http://www.dectech.org/publications/LinksNick/Language/Two%20wrongs%20make%20a%20right%20Learnability%20and%20word%20order%20consiste.pdf

Thorne, Steven L. (2000) Second language acquisition theory and the truth(s) about relativity. In James Lantolf (ed.), Sociocultural Theory and Second Language Learning. Oxford: Oxford University Press, 219-243.

Beockx, Cedric

Free Word Order in Minimalist Syntax. University of Maryland. Online available at http://209.85.165.104/search?q=cache:YfOYaDfVojQJ:mercury.hau.ac.kr/kggc/Resources/Papers/Boeckx03_Scrambling.pdf+linguistic+scrambling,+free+word+order&hl=en&ct=clnk&cd=16&gl=us.

Kim, Jong-Bok. 2000. The Grammar of Negation: A Constraint-Based Approach. Stanford: CSLI Publications.

Kim, Jong-Bok and Jaehyung Yang. 2003a. ‘Korean Phrase Structure Grammar in Constraint-Based Grammar and Building a Syntactic Parser with the Linguistic Knowledge Building System, (in Korean). Korean Linguistics 20: 1-40.

Kim, Jong-Bok and Jaehyung Yang. 2003b. Korean Phrase Structure Grammar and Its Implementations into the LKB System. Paper presented at the 17th Pacific Asia Conference on Language, Information, and Computation, October 1a€”3, 2003.

Jong-Bok Kim and Jaehyung Yang. Projections from Morphology to Syntax in the Korean Resource Grammar: Implementing Typed Feature Structures. Lecture Notes in Computer Science Vol.2945: 13-24, (Paper presented in CICLING 2004 Conference, Seoul, Korea), Springer-Verlag, 2004.2

Jong-Bok Kim, Jaehyung Yang and Incheol Choi. Feature Unification and Constraint Satisfaction in Parsing Korean Case Phenomena. Lecture Notes in Artificial Intelligence Vol.3339, pp.1160-66, (Paper presented at AI 2004: Advances in Artificial Intelligence), Springer, Dec 2004.

Jong-Bok Kim, Jaehyung Yang, Incheol Choi, “Capturing and Parsing the Mixed Properties of Light Verb Constructions in a Typed Feature Structure Grammar,” Proc. Of PACLIC 18: The 18th Pacific Asia Conference on Language, Information and Computation, pp.81-92, Waseda Univ, Tokyo, Dec 2004.

Jong-Bok Kim, Jaehyung Yang, “Parsing Mixed Constructions in a Typed Feature Structure Grammar,” Lecture Notes in Artificial Intelligence, Vol.3248, pp.42-51, Springer, Feb 2005.

Jong-Bok Kim, Jaehyung Yang, “Processing Korean Case Phenomena in a Typed-Feature Structure Grammar,” Lecture Notes in Computer Science, Vol.3406, pp.60-72, (Paper presented at CICLING-2005, Mexico City, Mexico), Springer, Feb 2005.

Jong-Bok Kim, Jaehyung Yang, “Parsing Korean Honorification in a Typed Feature Structure Grammar,” CLS (Chicago Linguistic Society) 41, (to appear), 2005.

Linguistics: Free Word Order, Scrambling – a Review