Juwon Lee

https://sites.google.com/view/juwonlee19/  


1. The Lexical Meaning of Tasi 'Again' 

I show that the Korean adverb tasi 'again' has a new type of repetitive reading, which I call pseudo-repetitive reading, and the previous lexical or structural accounts of again cannot explain the new reading. As a solution to this problem, I propose two lexical entries of tasi 'again', the hypernymic tasi and the antonymic tasi, and argue that both entries can derive the repetitive readings (including the pseudo-repetitive reading) and restitutive readings, respectively. This account of the semantic phenomena shows an important interaction between a semantic taxonomy and the lexical meaning of the adverb in Korean. In addition, I show that there are some similar data in English which suggest that a semantic taxonomy-based analysis is also needed for again in this language. In future work, I would like to investigate whether other languages also allow pseudo-repetitive readings of their corresponding adverbs.  
       
Lee, Juwon. 2017. Semantic Taxonomies and Tasi ‘Again’ in Korean. The Linguistic Association of Korea Journal 25.4, 61-83.

Lee, Juwon. 2018. Intentionality, Purpose, and Tasi ‘Again’ in Korean. The Linguistic Association of Korea Journal 26.2, 147-173.    

2. The Complementarity of Intentionality and Affectedness (CIA)

One of the primary purposes of linguistics is to specify what is possible and what is not possible in natural languages (see e.g. the issue of manner/result complementarity in Rappaport Hovav & Levin, 2010; Beavers & Koontz-Garboden, 2012). In this research, I propose the Complementarity of Intentionality and Affectedness (CIA) (see Lee, 2016b; Beavers & Lee, Accepted), namely that intentionality and affectedness cannot be entailed in a minimal accomplishment predicate (i.e. an accomplishment verb plus only its complement) at the same time. This generalization has been supported with Korean transitive verb constructions (Lee, 2015) and English conative alternations (Lee, 2016b). I am planning to test whether the CIA can be applied to various languages: non-culminations in e.g. Chinese and Japanese and conative alternations in e.g. Finnish and Warlpiri and I believe that the CIA could offer some important insights into what possible and impossible minimal predicates of accomplishments are in natural languages.        

Beavers, John & Juwon Lee. In press. Intentionality, Scalar Change, and Non-culmination in Korean Caused Change-of-State Predicates. The Special Issue of Linguistics on Non-culmination and Agent Control.   

Beavers, John and Andrew Koontz-Garboden. 2012. Manner and Result in the Roots of Verbal Meaning. Linguistic Inquiry 43.3: 331-369.

Lee, Juwon. 2015. An Intention-based account of accomplishments in Korean. Ph.D. dissertation, The University of Texas at Austin. 

Lee, Juwon. 2016b. Intentionality and Conative Constructions in English. Studies in Linguistics 41: 327-356.  

Lee, Juwon. 2018. The Complementarity of Intentionality and Affectedness: Evidence from English Verbs of Killing. Language and Information 22.3, 1-22. 

Rappaport Hovav, Malka, and Beth Levin. 2010. Reflections on Manner/Result Complementarity. In Edit Doron, Malka Rappaport Hovav, and Ivy Sichel (eds.), Lexical Semantics, Syntax and Event Structure, 21–38. Oxford: Oxford University Press.

3. Lexical aspect: activity vs. accomplishment

Non-culmination readings of causative accomplishment predicates have been much discussed in the literature, but other types of predicates have been little studied in relation to non-culmination. In this research, I discuss the relation between non-culmination and activity predicates in Korean (Lee, 2016a). Particularly, I hypothesize that some Korean predicates which have been regarded as activity predicates are in fact accomplishment predicates of some kind having a causative event structure, unlike English counterparts. As evidence for this recategorization, the failed attempt (FA) readings of “activity” predicates are discussed, along with other kinds of aspectual evidence. First, the FA interpretation of a predicate is possible only if the predicate has a causative event structure in Korean (Lee, 2015). Based on this generalization and the fact that the so-called "activity" predicates allow FA readings (e.g. Tom-i ttwi-ess-ciman, ttwi-l swu eps-ess-ta. (lit.) ‘Tom jumped, but he could not jump’ = ‘Tom tried to jump, but he could not jump’), I argue that they should actually belong to accomplishment (Lee, 2016a). I suggest then that accomplishments are classified into two types, state-accomplishment (e.g. kkay- ‘break’ and yel- ‘open’) and activity-accomplishment (e.g. ttwi- ‘jump’ and ket- ‘walk’). I show that this causative event structure of “activity” predicate can explain various facts regarding aspectual properties of the predicates, the ambiguity with keuy ‘almost’, the ambiguity with maney-adverbial (in-adverbial), and even the non-ambiguity with tasi ‘again’.

Lee, Juwon. 2015. An Intention-based account of accomplishments in Korean. Ph.D. dissertation, The University of Texas at Austin. 

Lee, Juwon. 2016a. Some Activity Predicates as Accomplishments. Language and Information 20.2, 117-143.   

4. Evidentiality  

The direct evidential -te in Korean has been much studied (see e.g. Sohn, 1994; Cinque, 1999; Chung, 2006; Lim, 2011; Lee, 2011). I discuss the complex interactions between the direct evidential -te, person, and experiencer predicates in Lee (2012b); particularly, I investigate how exactly the three components are related with each other in the evidential sentences, and how the interactions of the three components can be formally analyzed to correctly license only well-formed evidential sentences. I show that in Korean direct evidential construction with a non-private predicate (e.g. pwutulep- ‘soft’), the asserter (i.e. the speaker na ‘I’ in declarative or the addressee ne ‘you’ in question) must be the experiencer (i.e. the subject) of the predicate, but there is no such constraint in direct evidential construction with a private predicate (e.g. aphu- ‘sick’). I also argue that the direct evidential construction with a non-private predicate is an instance of self-ascription (see e.g. Wechsler, 2010, 2012 for self-ascription). Then I propose a formal analysis of the direct evidential sentences in the Minimal Recursion Semantics (MRS) (Copestake et al., 2005). Recently, I am working on interactions of evidentiality and non-culmination: if the past tense morpheme -ess is attached to the accomplishment verb in a direct evidential sentence, the potential FA reading is not allowed. My current hypothesis is that -ess locates the event described by the verb before the time when direct evidence obtains and so an occurrence of the inherent result is necessary to be directly observed.

Chung, Kyung-Sook. 2006. Korean Evidentials and Assertion. In Proceedings of the 25th West Coast Conference on Formal Linguistics, 105-113. Somerville, MA: Cascadilla Proceedings Project. 

Cinque, Guglielmo. 1999. Adverbs and Functional Heads: A Cross-Linguistic Perspective. Oxford: Oxford University Press.

Lim, Dong Sik. 2011. Evidentials in Interrogatives: A Case Study of Korean. Doctoral Dissertation, University of Southern California.

Lee, Juwon. 2012b. The Direct Evidential -te in Korean: It’s Interaction with Person and Experiencer Predicates. In Proceedings of the 19th International Conference on Head-Driven Phrase Structure Grammar, 251-271. Stanford: CSLI Publications.  

5. Coordination 

I have dealt with novel data of Korean NP-coordinations in which a non-final conjunct is semantically incompatible with the main verb (e.g. Jane-i pap-kwa khephi-lul masi-ess-ta. (lit.) ‘Jane drank the rice and the coffee.’ = ‘Jane ate the rice and drank the coffee.’) (Lee, 2015, In prep). This phenomenon raises the semantic issue of how to derive the appropriate verbal meanings related to the conjuncts incompatible with the main verbs. This is of empirical and theoretical importance, since to my knowledge this kind of data has not been discussed before, and the syntax-semantics interface involved in the phenomenon poses significant challenges to prior analyses of coordinations. I propose the hypothesis that some hypernym or sister of the main verb in semantic taxonomy is selected as the verbal meaning for the incompatible conjunct in the construction.   

Lee, Juwon. 2015. Abstract. Semantic Taxonomies and NP-coordinations. Talk at the 22nd International Conference on Head-Driven Phrase Structure Grammar. 

6. Light Verb 

Korean light verb constructions (LVCs) (e.g. Jane-i pap-ul hay-ss-ta. (lit.) ‘Jane did the rice.’ = ‘Jane cooked the rice.’) pose a significant problem to compositionality (Lee, 2011a, 2011b, Under review). These LVCs also have the typical problems of multiword expressions (MWEs) (Sag et al., 2002). First, the word-with-space approach would result in a proliferation in the lexicon, since each combination of a common noun (CN) and the light verb ha- 'do' (e.g. pap-ul ha- 'rice-Acc do', khephi-lul ha- 'coffee-Acc do', etc.) should be listed in the lexicon as a single lexeme. This is an undesirable redundancy, missing a generalization. Second, a compositional analysis (e.g. a syntactic rule combining the light verb ha- 'do' with a CN object) could overgenerate ungrammatical LVCs (e.g. *Jane-i mwul-ul hay-ss-ta. (lit.) ‘*Jane did the water.’), since it is not clear why a certain CN like mwul ‘water’ is disallowed in the LVC (lexical idiosyncrasy). In this research, I analyze various Korean transitive verb constructions whose interpretations are associated with some qualia roles (see Pustejovsky 1995 for the notions of qualia roles) (Lee, Under review). I investigate how exactly the related meanings of the transitive verb constructions arise, employing the notions of qualia roles and feature structures. In particular, I argue that the qualia roles of a CN are related to each other in certain ways (e.g. there is a ranking among some qualia roles of a CN) rather than simply being listed in the qualia structure.

Lee, Juwon. 2011a. Two Types of Korean Light Verb Constructions in a Typed Feature Structure Grammar. In Proceedings of the 2011 Association for Computational Linguistics Workshop on Multiword Expressions: from Parsing and Generation to the Real World, 40-48. 

Lee, Juwon. 2011b. Passive Light Verb Construction in Korean: Lexical Representation of toy. Harvard Studies of Korean Linguistics 14, 25-45. 

Lee, Juwon. 2016c. Lexicology of Korean Common Nouns. Korean Journal of Linguistics 41.4, 715-739.    

7. Serial Verb 

Serial verb construction (SVC) is a structure consisting of more than two component verbs but denotes what is conceptualized as a single event. SVC is an important part of the study of complex predicates. A central issue of SVC is how the arguments of the component verbs of an SVC are realized in a sentence. In the literature, it is generally assumed that the constituent verbs of an SVC share the subject (Foley and Olson 1985, Sebba 1987, Lee 1992, Andrews 1997, Chung and Kim 2008, Müller and Lipenkova 2009, Kim 2010, among others) or they share the object (Baker 1989) or an internal argument (e.g. themes, instruments, goals) (Collins 1997). I show that the novel data of Korean passive SVCs violate all the subject-sharing, object-sharing, and the argument saturation and then I propose two hypotheses: i) Korean SVCs are broadly classified into two types, subject-sharing SVCs where the subject is shared by the verbs and index-sharing SVCs where only indices of semantic arguments are shared by the verbs and ii) an argument index sharing is a general requirement of SVCs in Korean. I also argue that an argument composition analysis can accommodate the Korean passive SVCs straightforwardly compared to other alternative analyses.

Lee, Juwon. 2012a. Change of State Verb and Syntax of Serial Verb Constructions in Korean: An HPSG Account. In Proceedings of the Poster Session of the West Coast Conference on Formal Linguistics, Coyote Papers: Working Papers in Linguistics 20, 57-65.

Lee, Juwon. 2014a. Two Types of Serial Verb Constructions in Korean: Subject-Sharing and Index-Sharing. In Proceedings of the 21st International Conference on Head-Driven Phrase Structure Grammar, 136-155. Stanford: CSLI Publications.

Lee, Juwon. 2014b. Multiple Interpretations and Constraints of Causative Serial Verb Constructions in Korean. In Proceedings of the 38th Annual Meeting of the Berkeley Linguistics Society, 288-306. 

8. Resultative

Lee (To appear) discusses what is referred to as predicative resultative constructions in Korean (e.g. Tom-i thakca-lul kkaykkusha-key takk-ass-ta. ‘Tom wiped the table clean.’) and argues that they are in fact a kind of clausal resultative construction (see the different types of resultatives in Wechsler and Noh, 2001). While I agree with the general idea that the resultative predicate forms a clause in Korean (Shim & den Dikken, 2007; Shibagaki, 2011), particularly I propose the following hypotheses: (i) the resultative predicate, X-key, is morpho-syntactically an adverb rather than an adjective, (ii) X-key forms a fully saturated clause, result clause (sometimes with the predication subject omitted), and (iii) the result clause is a complement of the main verb. I then propose a unified analysis of the resultative constructions. I am also working on the complex relations between resultative predicate, negation and intentionality. In English it is impossible to add the negation not to a resultative predicate (e.g. *John wiped the table not clean), but it is possible in Korean (see e.g. Son, 2008) and the intentionality on the part of the subject arises from the negative resultative predicate, which has been rarely discussed in the literature. 

Lee, Juwon. 2016d. Towards a Unified Account of Resultative Constructions in Korean. In Proceedings of the 30th Pacific Asia Conference on Language, Information and Computation (PACLIC), 501-510. 

Lee, Juwon, Eunjeong Oh, and Sanghoun Song. 2018. An Experimental Study of Resultative Constructions with X-key. Studies in Linguistics 49, 249-274.

Lee, Juwon, Eunjeong Oh, and Sanghoun Song. 2018. A Clausal Analysis of Resultative Constructions with X-key: Evidence from Corpus and Experiments. Language and Information Society 35, 207-244.