從優選理論分析英文縮詞與混合詞之音韻保留形式 - 政大學術集成

全文

(1)國立政治大學語言學研究所碩士論文. National Chengchi University Graduate Institute of Linguistics Master Thesis. 指導教授：蕭孙超 Advisor: Yuchau E. Hsiao. 立. 政治大. ‧ 國. 學. 從優選理論分析英文縮詞與混合詞之音韻保留形式. ‧. Phonological Preservation of English Clips and Blends: An Optimality-Theoretic. n. al. er. io. sit. y. Nat. Analysis. Ch. engchi. i n U. 研究生：林綠茜撰 Student: Lu-Chien Lin 中華民國九十九年六月 June, 2010. v.

(2) 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v.

(3) PHONOLOGICAL PRESERVATION OF ENGLSIH CLIPS AND BLENDS: AN OPTIMALITY-THEORETIC ANALYSIS. BY. 學. ‧ 國. 立. 政治大 Lu-Chien Lin. ‧. n. al. er. io. sit. y. Nat. A Thesis Submitted to the Graduate Institute of Linguistics in Partial Fulfillment of the Requirements for the Degree of Master of Arts. Ch. engchi. June, 2010. i n U. v.


(5) 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. Copyright @ 2010 Lu-Chien Lin All Rights Reserved iii. v.

(6) Acknowledgements 「當你真心想要一件事情，全世界都會來幫你。」－《牧羊少年奇幻之旅》. 三年前我用這句話勉勵重考的自己，最後終能如願以償進入語言學的研究領域。從碩一開始，每當我在研究上或生活上遇到挫折時，總會有人安慰我，鼓勵我，也有人會跟我一起想法子，透過好多貴人的幫忙，問題總能迎刃而解，也讓我對牧羊少年的話更深信不疑。三年的研究生涯充實而精采，首先，我要感謝我的指導教授蕭孙超老師，感謝您一路的栽培，無論是論文的指導或是做人處事的道理，我會永遠記得您的諄諄教誨，也感謝您總是鼓勵我投稿以及不厭其煩地為我批改摘要，讓我有機會在國際的學術會議上發表自己的研究，在未來繼續進修的日子裡，我會更努力以一顆年輕及充滿熱忱的心去做研究，就如同您總是對音韻學充滿熱情一般。想起您總是說我的運氣很好，人生順順利利的似乎沒有煩惱，但其實很多事情如果沒有您的提攜，我想我會跌跌撞撞的，所以說您是我的貴人真的是十分適切。同時，我要感謝我的口試委員政大英文系莫建清老師、清大語言所曹逢甫老師以及政大語言所何萬順老師，感謝三位老師對論文題目的稱讚與肯定，也感謝您們在計劃書答辨以及論文答辯時，提供了很多寶貴的意見讓整篇論文更臻完善。謝謝您們，辛苦了！. 立. 政治大. ‧. ‧ 國. 學. n. al. er. io. sit. y. Nat. 再來，我要感謝我在求學期間遇到的所有老師，包括溫柔美麗的黃瓊之所長、活潑亮麗的萬依萍老師、教學認真的徐嘉慧老師以及學富五車的詹惠珍老師，謝謝您們在課堂上的指導，使我的語言學知識及論文撰寫之先備知識更加豐富。在此也同時感謝日文系的葉寶珠老師與彭南儀老師的教學，讓我對日文的學習充滿了興趣。接下來我要特別感謝政大音韻幫：熊芮華、粘復真、凌旺楨、王麗婷、童啟美、黃子權以及許淨涵。因為有了你們的情義相挺，使我的研究之路充滿了歡笑，也謝謝你們與我一同籌備貓空論壇，讓這場音韻學研討會圓滿落幕。另外，我也感謝我的同班同學們三年來的分享與關心，更感謝所有鼓勵我的學長姐以及惠玲. Ch. engchi. i n U. v. 助教，每當我對求職或考試或行政程序上有不懂的地方，都因為你們的熱心協助而獲得解決的方式。當然，我要感謝生命中最重要的家人：父親林聰明先生、母親江秀梅女士以及我可愛的妹妹與弟弟。謝謝爸媽在生活上與經濟上給予我最大的支持，讓我的求學之路不曾為了錢而煩惱奔波。謝謝弟弟妹妹一路來為我加油打氣，大多時間還把家裡的電腦讓給我，讓我順利完成論文初稿的撰寫。最後，謝謝我的開心果梁益昌先生，因為你的搞笑，我的負面情緒總能消失得很快，能與你相識，我覺得很幸福，所以我胖了！對了，我也要謝謝我最寶貝的寵物兔養樂多，謝謝你三年來健健康康的，沒有讓我擔心，希望你能繼續活蹦亂跳地陪伴著我。. iv ii.

(7) TABLE OF CONTENTS. CHAPTER 1 INTRODUCTION.......................................................... 1 1.1. Background ............................................................................................ 1. 1.2. Research Questions ................................................................................ 3. 1.3. Thesis Organization ............................................................................... 4. 政治大 Optimality Theory .................................................................................. 5 立. CHAPTER 2 LITERATURE REVIEW.............................................. 5 2.1. 2.1.5. Cophonology Theory ................................................................... 12. ‧. ‧ 國. Basic Concepts of Classic OT........................................................ 5 Correspondence Theory ................................................................. 8 The schema of Output-to-Output Correspondence ...................... 10 Anchoring .................................................................................... 10. 學. 2.1.1 2.1.2 2.1.3 2.1.4. y. Nat. 2.3. Blending as Word-Formation ............................................................... 14. 2.4. Bat-El (1996) ....................................................................................... 16. 2.5. Lappe (2008) ........................................................................................ 19. 2.6. Summary .............................................................................................. 20. n. al. er. sit. Clipping as Word-Formation ............................................................... 14. io. 2.2. Ch. engchi. i n U. v. CHAPTER 3 ENGLISH CLIPPING: AN OT ANALYSIS ............. 23 3.1. An OT Account for Morphological Clipping ....................................... 24. 3.2. An OT Account for Phonological Clipping ......................................... 29 3.2.1 3.2.2. Left-edge Preservation ................................................................. 31 Right-edge Preservation ............................................................... 46. 3.3. A Cophonology Theory for Clipping Variations .................................. 50. 3.4. Summary .............................................................................................. 57 v.

(8) CHAPTER 4 ENGLISH BLENDING: AN OT ANALYSIS ........... 59 4.1. An OT Account for Morphological Blending ...................................... 60. 4.2. An OT Account for Phonological Blending......................................... 64. 4.3. Left-Anchoring Constraints in a Stringency Relation.......................... 73. 4.4. A Cophonology Theory for Blending Variations ................................. 75. 4.5. Cross-Linguistic Evidence: Spanish Blending..................................... 81. 4.6. 4.5.1. Spanish Blends ............................................................................. 81. 4.5.2. Constraints and Rankings ............................................................ 83 Summary .............................................................................................. 87. CHAPTER 5. 治政 CONCLUSION ............................................................ 89 大立 ‧. ‧ 國. 學. References ............................................................................................... 93. n. er. io. sit. y. Nat. al. Ch. engchi. vi ii. i n U. v.

(9) 國. 立. 政. 治. 大. 學. 研. 究. 所. 碩. 士. 論. 文. 提. 要. 研究所別：語言學研究所論文名稱：從優選理論分析英文縮詞與混合詞之音韻保留形式指導教授：蕭孙超研究生：林綠茜論文提要內容：(共一冊，17185 字，分五章十九節). 政治大. 本文以優選理論的觀點分析英文縮詞與混合詞的音韻保留形式。音韻保留形. 立. ‧ 國. 學. 式分為：來源詞之聲母保留、來源詞之單音節保留以及來源詞之雙音節保留。本研究認為英文縮詞與混合詞的音韻保留策略不只一種，應用不同的策略會產生不. ‧. 同類型的縮詞及混合詞，筆者透過並存音韻理論(Cophonology Theory)來說明英. sit. y. Nat. io. n. al. er. 文縮詞與混合詞的音韻保留策略是多個次語法的運作結果。縮詞的音韻保留策略. i n U. v. 有四種，可分為左邊保留及右邊保留，其中又以左邊保留佔多數，在這兩種保留. Ch. engchi. 中又分別有兩種模組(template)保留策略。混合詞的音韻保留策略有三種，主要由 MAXS2 這條可移動制約的位置來決定，當它在層級中移動到不同的位置會形成不同的保留策略。此外，本文也提供了跨語言分析，發現西班牙混合詞與英文混合詞可由相似的制約透過不同排序來解釋，表示不同語言的混合詞，其行為相當類似。簡言之，本篇論文藉由優選理論的觀點，首度就英文縮詞與混合詞提出了一個整體分析。. vii.

(10) Abstract. This thesis examines the nature of English clipping and blending from the perspective of Optimality Theory. Clipped and blended words may use phonological strategies to preserve part of the source such as the preservation of the onset, syllable, or foot. Different strategies of preservation form different patterns of clipped or blended words. This thesis illustrates that these phonological strategies are determined. 政治大. by the different cophonologies. There are four strategies in forming clipped words.. 立. Clipped words can be preserved from the left edge or the right edge of the source,. ‧ 國. 學. each of which follows either a bimoraic template or a disyllabic template. There are. ‧. three strategies in forming blended words, depending on the ranking of the. Nat. io. sit. y. unspecified constraint MAXS2. In addition, the present study offers cross-linguistic. er. evidence from Spanish blends, showing that Spanish blending and English blending. al. n. v i n C hthis thesis has provided share certain similarities. To conclude, e n g c h i U a theoretical. generalization of English clipping and blending, taking a constraint-based approach.. viii ii.

(11) CHAPTER 1 INTRODUCTION. 1.1 Background This thesis employs a constraint-based theory to examine the nature of two types of English word-formations: clipping and blending. A clipped word is the shortened part of an existing word, like phone from telephone. A blend is the derived part in the combination of two or more words, like brunch from breakfast and lunch. There are. 治政 two goals in this research. The first is to describe the 大divisions of clipped and blended 立 ‧ 國. 學. data into several types in terms of the preservation of the first source word. The second is to account for these types under the framework of Optimality Theory (OT).. ‧. Specifically, this study provides a Classic OT account and a Cophonology (extension. sit. y. Nat. io. al. n. any given type.. er. of OT) explanation for the mapping between the source and the preserving output of. Ch. engchi. i n U. v. The morphological procedures of clipping and blending have been discussed in the literature of linguistics (Adams 1973, Algeo 1977, 1978, Bauer 1983, Cannon 1986, Kelly 1998, Berman 1989, Bat-El 1996, Minkova and Stockwell 2001, Kemmer 2003, Gries 2004). Algeo (1978) suggests that they be regarded as formal word-making devices in the language. Adams (1973) defines clipping as a process that shortens a word of two or more syllables, without changing its function. Bauer (1983) also offers a definition of clipping, as in the following: 1.

(12) 2. “Clipping refers to the process whereby a lexeme (simple or complex) is shortened, while still retaining the same meaning and still being a member of the same form class. (Bauer 1983)” From Bauer’s perspective, a clipped output is the shortened form of an existing word, which is identical to its source syntactically and semantically. In addition, Bauer (1983) defines a blend as a new morphological item formed from the combination of parts of two or more words, which is different from but related to its sources semantically, with the meaning of the blend originating from the integration of the. 政治大 meanings of the sources. According to these definitions, we find that the processes 立. ‧ 國. 學. involved in clipping and blending are similar in word-formation, where part of the. ‧. existing word is preserved in the creation of a new word. This present research. sit. y. Nat. focuses on the analysis of the segmental preservation in clipping and blending on the. n. al. er. io. basis of Bauer’s definitions. Segmental preservation is a type of morphological. Ch. i n U. v. construction that contains a part identical to the existing word. The existing word is. engchi. considered the source and the preserved part is referred to as the clipped output. The combination of two or more clipped outputs forms a blended output. These shortened forms will be further studied in the following discussions. Several observations bring about this present study. First, both clipping and blending preserve parts of the existing words to create a new word. Second, there is a positional correspondence between the source and the preserved output. Finally, the.

(13) 3. clipped words and blended words are both subject to specific templates. We find that a majority of clipped outputs occur in a structure with a monosyllabic or disyllabic template, and the size of most blends is identical to that of the second source. In brief, clipping and blending are similar processes in the formation of a new word. 1.2 Research Questions The present study is of interest in that a clipped output or a blended output is not. 政治大. necessarily a morpheme of the source. It discovers that there are several phonological. 立. strategies governing these morphological processes. First, both processes may. ‧ 國. 學. preserve the first syllable of the source. Second, both processes may preserve the first. ‧. syllable and the second onset of the source. Finally, both processes may preserve the. Nat. io. sit. y. first two syllables of the source. In this study, these strategies are investigated from. n. al. er. the point of view of the constraint-based theory and the following issues are discussed:. Ch. engchi. i n U. v. a. Are the constituents preserved randomly in clipping and blending? b. What kinds of constituents of the source are preserved in clipping and blending? c. Under the model of OT, how is the preservation of constituents be explained and how are variations analyzed?.

(14) 4. 1.3 Thesis Organization The thesis is organized as follows: Chapter 1 gives an outline of the current study, raising several research questions as above. Chapter 2 introduces earlier research pertinent to our discussion, and reviews the theoretical frameworks used in this thesis. Chapter 3 examines the nature of English clipping under an OT schema. Also, in Chapter 4, OT is used to account for the nature of English blending and offers. 政治大. cross-linguistic evidence. Chapter 5 concludes this study.. 立. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v.

(15) CHAPTER 2. LITERATURE REVIEW. 2.1 Optimality Theory. 2.1.1. Basic Concepts of Classic OT. Optimality Theory (OT) is a constraint-based framework which is proposed by Prince and Smolensky (1993, 2004). Different from the traditional model of The. 政治大. Sound pattern of English (Chomsky and Halle, 1968), the OT framework does not. 立. develop on the basis of ordered rules1. Rather, it advocates the use of a set of. ‧ 國. 學. constraints to govern the grammar of language. Different languages are characterized. ‧. by different rankings of the constraints. The main idea of OT expresses that a surface. Nat. io. sit. y. output results from competition against possible candidates. The competition is. er. fulfilled by means of three components under the framework of OT, including. al. n. v i n CONSTRAINT, GENERATORC , and . The three components are h eEVAL n gUATION chi U responsible for the operation of OT, as indicated in (1).. 1. For more details about the development of phonology from 1970s to 1990s, see Archangeli (1997) and McCarthy (2008). 5.

(16) 6. (1) OT schema Input. CONSTRAINT CON 1. GENERATOR Candidate (a) Candidate (b) Candidate (c) :. CON 2. CON 3. CON 1 >> CON2 >> CON3. Output. EVALUATION. Given an input in the OT schema (1), component GEN associates it with possible. 政治大. representations. These representations are referred to as candidates. CON contains a. 立. set of ranked constraints. The ranking is used to evaluate the possible candidates.. ‧ 國. 學. There is no intermediate stage for evaluation within the model. Candidates go through. ‧. a parallel evaluation of the ranking of the constraints and the constraint ranking. y. Nat. er. io. al. sit. selects the most harmonic one as the optimal output.. n. In OT, constraints are universal, violable, and ranked language-particularly.. Ch. engchi. i n U. v. Specifically, the universality of the constraints means that the same set of constraints is used to construct grammar in all languages. These constraints are ranked in different ways, depending on the nature of a given language. In addition, these constraints can be violated. The violation of a lower-ranking constraint can be allowed in order to fulfill the requirement of a higher-ranking constraint. The optimal output is selected if there is no violation or only the minimal violations of the constraints or by the satisfaction of a higher-ranking constraint. A formal OT approach to the choice of.

(17) 7. the surface output is generalized as follows: (2) CONSTRAINT 1 >> {CONSTRAINT 2, CONSTRAINT 3} >> CONSTRAINT 4 /Input/. CON 1. CON 2. CON 3.  Candidate (a). *. Candidate (b). *!. Candidate (c) Candidate (d). CON 4. *! *!. Conventionally, the evaluation of OT is illustrated by tableaux, as shown in (2). The. 政治大 representations. The top立 row lists relevant constraints and forms a ranking. Constraint top left-most cell presents an input, which is longitudinally followed by its possible. ‧ 國. 學. 1 ranks above Constraint 2 and Constraint 3, which dominate Constraint 4. The finger. ‧. marker indicates the best choice of the evaluation. The solid line distinguishes the. sit. y. Nat. priority accorded to one constraint over another. The left-side constraint takes a higher. al. er. io. priority than the right-side constraint if they are separated by a solid line. The dotted. n. v i n line between two constraintsC shows that the ranking is unknown. The asterisk denotes hengchi U a violation and the exclamation mark represents a fatal violation that rules out a losing candidate. When a candidate is fatally removed, the cells of lower-ranking constraints are shaded. The shading of cells means that the constraints are irrelevant to the competition. As Tableau (2) shows, candidate (d) is first ruled out because it violates the highest-ranked constraint. Candidate (b) and (c) are equally less harmonic because.

(18) 8. they violate the constraints at the same level of the ranking. Although candidate (a) also incurs a constraint violation of the lowest ranked constraint, it satisfies the three higher-ranking constraints and thus emerges as the optimal output.. 2.1.2 Correspondence Theory. The description of OT provided above concentrates on the mapping of the relationship between a given input and its surface output by means of faithfulness. 治政 constraints, which regulate the consistency between two strings. 大 The use of 立 ‧ 國. 學. faithfulness constraints is extended by McCarthy and Prince (1994, 1995). They propose Correspondence Theory, using faithfulness constraints to examine the. ‧. relationship between the reduplicant and its base. Benua (1995) refers to the relation. sit. y. Nat. io. n. al. er. between two outputs as Output-to-Output correspondence. The main idea of. i n U. v. Correspondence Theory is that correspondent elements, including input-to-output (IO). Ch. engchi. relation and base-to-reduplicant (BR) relation, should be governed by faithfulness constraints. Thus, the family of faithfulness constraints has to be more specific to account for IO relation and BR relation, as in (3)-(5) proposed by McCarthy and Prince (1995)..

(19) 9. (3) The MAX Constraint Family (Specific Instantiations) MAX-BR: Every segment of the base has a correspondent in the reduplicant. (Reduplication is total.) MAX-IO: Every segment of the input has a correspondent in the output. (No phonological deletion.) (4) The DEP Constraint Family (Specific Instantiations). 政治大. DEP-BR: Every segment of the reduplicant has a correspondent in the base.. 立. (Prohibits fixed default segmentism in the reduplicant.). ‧ 國. 學. DEP-IO: Every segment of the output has a correspondent in the input.. ‧. (Prohibits phonological epenthesis.). Nat. io. sit. y. (5) The IDENT(F) Constraint Family (Specific Instantiations). al. er. IDENT–BR(F): Reduplicant correspondents of a base [(F] segment are also [(F].. n. v i n Ch IDENT–IO(F): Output correspondents i U [(F] segment are also [(F]. e n gofc anh input. Constraints MAX-IO, DEP-IO, and IDENT-IO regulate the correspondence between the input and the output. They prohibit the deletion, addition, or change of a given input. Constraints MAX-BR, DEP-BR, and IDENT-BR guard the correspondence between outputs (base/ reduplicant). They protect the base from being deleted, inserted, or changing. Thanks to the Correspondence model, the relation between surface forms can be examined by OT schema..

(20) 10. 2.1.3 The schema of Output-to-Output Correspondence. Like the analysis of reduplication, Benua (1995) employs Correspondence Theory to investigate truncation, presenting a schema of output-to-output correspondence in (6). (6) Truncation schema (Benua 1995) BT-Identity. 學. IO-FAITH. ‧ 國. Base. 治政 - - - - - - - - - - - Truncated Form 大立 ‧. Input. Nat. io. sit. y. The schema illustrates the mapping relation between the truncated form and its base.. al. er. It characterizes truncation as two types of correspondent relations: an input-to-base. n. v i n Coutput-to-output mapping (IO-correspondence) and an U (OO-correspondence) h e n g c h imapping between the base and the truncated form. To maintain BT-Identity, the constraints MAX-BT, DEP-BT, and IDENT-BT are relevant to govern the truncatory process.. 2.1.4 Anchoring. Anchoring is another kind of faithfulness constraint which regulates the identity on the edges of two correspondent strings. It demands that the two strings must share.

(21) 11. the same constituent at the designed periphery, proposed by McCarthy and Prince (1995), as shown in (7): (7) {RIGHT, LEFT}-ANCHOR(S1 , S2) (McCarthy & Prince, 1995) Any element at the designated periphery of S1 has a correspondent at the designated periphery of S2 Let Edge(X, {L, R}) = the element standing at the Edge = L,R of X. RIGHT-ANCHOR . If x=Edge(S1 , R) and y=Edge(S2 , R) then xRy. LEFT-ANCHOR. Likewise, mutatis mutandis. Left-Anchor requires the completeness of the constituent at the left edge of two. 治政 independent relevant outputs. Right-Anchor demands 大that the constituent at the right 立 ‧ 國. 學. edge of the output is identical to that of the base. Previous studies show that anchoring constraints often have an effect on reduplication. Marantz (1982) suggests that there is. ‧. a tendency for locality between the reduplicant and the base: a reduplicant is placed as. sit. y. Nat. io. n. al. er. close as possible to its correspondent in the base. Thus, prefixing reduplicants. i n U. v. correspond to the initial part of the source and suffixing ones correspond to the final. Ch. engchi. part of the source. McCarthy and Prince (1995) point out that the tendency for locality is motivated by anchoring constraints. The ranking L-Anchor >> R-Anchor accounts for prefixing reduplication while R-Anchor >> L-Anchor explains suffixing reduplication. Nelson (2003) reveals an asymmetric relation between Left-Anchor and Right-Anchor and concludes that only left edge anchoring constraints are necessary. Lunden (2004) notes that both anchoring and alignment constraints are needed to account for the few cases which do not follow the tendency of locality. In this thesis,.

(22) 12. anchoring constraints are employed to account for the preservation of the constituent of the source.. 2.1.5 Cophonology Theory. In OT, there is a single fixed ranking of the constraints for each language. That is, constraints cannot be re-ranked in a language. If the ranking of the constraints is reordered, it composes the grammar of another language. However, on the. 治政 cophonology approach (Inkela and Zoll, 2007; Anttila and 大 Cho, 1998; Orgun, 1996; 立 ‧ 國. 學. Ito and Mester, 1993, 1999), the phonology of a language can consist of co-existing distinct subgrammars if the language shows different phonological patterns. Anttila. ‧. (1997, 2002) proposes the partial ordering model to account for cophonological. sit. y. Nat. io. n. al. er. categories. Under this model, the grammar of a language is not completely determined.. i n U. v. Certain unspecified constraints are ranked in different ways depending on the nature. Ch. engchi. of the subgroup of the grammar, as formalized in (8)..

(23) 13. (8) The formulation of a grammar lattice (Anttila, 2002)  {ABC, ACB, BAC, BCA, CAB, CBA} (Universal grammar) A>>B {ABC, ACB, CAB} (Language L) A>>B A>>C {ABC, ACB} (Subgrammar 1). 學. A>>B C>>B A>>C {ACB}. A>>B C>>B C>>A {CAB} (Subgrammar 5). al. er. io. sit. y. (Subgrammar 4). Nat. (Subgrammar 3). ‧. A>>B A>>C B>>C {ABC}. ‧ 國. 立. 政治大. A>>B C>>B {ACB, CAB} (Subgrammar2). n. Given three constraints, there are six rankings to consist of the grammar. There is no. Ch. engchi. i n U. v. further designated ranking in the total orderings (ABC, ACB, BAC, BCA, CAB, and CBA). When a ranking is designated (A>>B), the grammar that makes up Language L becomes more specific. From a cophonological perspective, constraints can be reordered within a language. This thesis examines English clipping and blending types in light of cophonologies of morphological grammar..

(24) 14. 2.2 Clipping as Word-Formation Adams (1973) has discussed the situation in which clipped words are used, finding that they are commonly expressed in less formal situations than their source. The informal usage can reflect the speaker’s familiar attitude toward the listener or to the object to which reference is being made. Adams observes, however, that no clear phonological rules govern the way in which the source is cut. Bauer (1983) agrees. 政治大. that the clipped form is shorter than its source while it is not known how many. 立. syllables of the source are retained. He doubts whether clipping is phonologically. ‧ 國. 學. predicted. Minkova and Stockwell (2001) point out that the existence of clipped. ‧. words can save time and space; however, they do not explain the grammar of clipping. Nat. io. sit. y. in more detail. These previous studies take a number of English clipped examples but. n. al. er. they do not provide a systematic account for the operation of clipping, especially in the area of phonology.. Ch. engchi. i n U. v. 2.3 Blending as Word-Formation In English, blending is another important mechanism to create new lexemes. In fact, it is a productive morphological device that exists in many languages of the world (Cannon 1986, Minkova and Stockwell 2001, Kemmer 2003, Gries 2004). Cannon (1986) reviews various terms about and definitions of blending, finding that blends have been referred to as haplology, jumbles, lapses, telescopes, and.

(25) 15. portmanteau words. In the literature, the definitions of a blend are not clearly created in a systematic way. The central focus of the studies above lies in a discussion of the terminology, definition, popularity, and usage of blends based on quantitative grounds. The following literature is concerned with the discussion of internal structure of the blend. Kelly (1998) assumes that the linguistic determinants that govern the. 政治大. structure of blends contain a coordinate structure and an orthographic factor. In spite. 立. of his assumption, many other studies have claimed that the structure of blends can be. ‧ 國. 學. explained more accurately by reference to their occurrence as a phonological. ‧. phenomenon. Algeo (1977) compares blending with other word formations,. Nat. deletion of overlapping of sounds, as in (9).. n. al. Ch. engchi. (9) The structure of blends (Algeo 1977) a.. er. io. sit. y. suggesting that the creation of blends is predictable from the partial omission or. i n U. v. They combine two or more forms, in which respect they are like compounds (hothouse) or affixed derivatives (hopeful).. b.. Some of them omit part of one source form, or parts of both, in which respect they are like stump words (exam).. The observations in (9) do not describe how these sources are truncated in a systematic way. Berman (1989) specifies the development of the blend in Hebrew as a.

(26) 16. fusion of stem + stem combinations. This finding may be language-particular. For instance, the English blend smog from smoke and fog is an exception to the combination of two stems. Kemmer (2003) presents a schema-based investigation of lexical blends, showing the possible range of form-meaning mapping between the blend and its sources. The schema-based is a constraint-based approach, providing an integrated and cognitive account for the conceptual structure of lexical blends. Gries. 政治大. (2004) discusses the contribution of the source in blend formation, finding that. 立. have an effect on forming a blend.. 學. ‧ 國. recognizability of the source and the similarities between the source and the blend. ‧. Nat. al. er. io. 2.4 Bat-El (1996). sit. shares highly phonological similarities with its sources.. y. Nevertheless, Kemmer and Gries’s studies do not clearly explain how the blend. n. v i n C h of the sourcesUin Hebrew blends is not Bat-E1 (1996) argues that the sequence engchi. unordered, as given in (10). (The underlined boldface stands for the preserved part of the source.).

(27) 17. (10) Hebrew blends (adapted from Bat-E1, 1996) Sources. Optimal blend. Ill-formed blend. mitnaxalim, mexablim mitnaxablim demokrat, diktator paxman, meiman. *mexablim. mitna(xalim)+(me)xablim. mexab(lim)+(mitnaxa)lim. demoktator. *dikrat. demok(rat) + (dik)tator. dik(tator) + (demok)rat. paxmeiman. *meiman. pax(man)+meiman. mei(man)+(pax)man. In (10), there are two sources for each blend. The left edge of the first source. 政治大 corresponds to the right立 edge of the blend. For the blend to be harmonic, the first. corresponds to the left edge of the blend. The right edge of the second source. ‧ 國. 學. source may consist of ‘mitnaxalim’, ‘demokrat’, or ‘paxman’ and the second source. ‧. may consist of ‘mexablim’, ‘diktator’, or ‘meiman’. Bat-E1 proposes that the order of. sit. y. Nat. the sources is determined by constraint interaction rather than the function of the. n. al. er. io. specific constraint which demands a sequence of the source elements, as shown below.. Ch. engchi. i n U. v. (11) Constraints on the number of syllables σMAX: Every syllable in both elements of the base must have a correspondent in the blend. (It would be violated as if syllables were deleted.) σDEP: Every syllable in the blend must have a correspondent in both elements of the base. (It would be violated as if syllables were added.).

(28) 18. (12) adapted from Bat-E1, 1996 / demokrat, diktator /.  a.. demok(rat+dik)tator 3σ 3σ. demoktator 4σ. b.. demokra(t+dik)tator 3σ 3σ. demokratator 5σ. c.. dik(tator+ demok)rat 3σ 3σ. dikrat 2σ. σMAX. σDEP * **!. *!. Tableau (12) considers two possible sequences of the sources. MAX and DEP pose. 政治大 sources. Hence, Bat-E1 proposes 立that the sources are not ordered randomly.. restrictions on the number of the syllables in the blend rather than on the order of the. ‧ 國. 學. Furthermore, it is not necessary to use a specific constraint that requires the order of. ‧. the sources. It can be indirectly determined by the interaction of independently. sit. y. Nat. motivated constraints. On the basis of Bat-E1’s proposal, I arrange the order of the. n. al. er. io. sources inferred from the structure of blended outputs in English. Specifically, the left. Ch. i n U. v. edge of the word is preserved in the blend is regarded as the first source, and the right. engchi. edge of the word is preserved in the blend is regarded as the second source. In Bat-E1’s study, σMAX and σDEP refer to syllable nodes instead of to the segmental content of the corresponding syllables. σMAX requires that the blend must not be smaller than the longest source. σDEP requires that the blend must not be larger than the longest source. Since σMAX outranks σDEP, the length of the blended output is never shorter than that of the longest source, as shown in (12) above. The.

(29) 19. property of the size of Hebrew blends is different from that of English blends. In English, the syllabic length of the blend is identical to that of the second source. For the template, I use the family of IDENT constraints or a specific markedness constraint to account for the size of the blend. MAX and ANCHOR are used to require the segmental preservation of the source. 2.5 Lappe (2008). 政治大. Lappe (2008) examines the structure of English truncated patterns on the basis of. 立. three levels of description. The first level is word structure, where truncated forms can. ‧ 國. 學. be described in terms of the number of syllables, stress assignment, and the part of the. ‧. base which is retained. The second level is syllable structure, where truncated forms. Nat. io. sit. y. are described in terms of syllabic constituents such as onset, nucleus, and coda. The. al. er. third is segmental level, which imposes restrictions on the kind of segments that make. n. v i n Cthat up the output. Lappe assumes h etruncatory i U differ from one another n g c hpatterns. according to the degree of constraints on different levels of structure. Each truncatory process is subcategrized in terms of the ranking of markedness and faithfulness constraints. Thus, Lappe shows that the difference between monosyllabic and disyllabic clippings is decided by the competition of the various specific MAX constraints. This paper divides English monosyllabic and disyllabic clipped data into several.

(30) 20. types and proposes ANCHOR constraints instead of MAX constraints to account for different types of clipped outputs. One reason for the adoption of ANCHOR constraints is that MAX is often used to regulate the segmental relation between the base and truncated output. It disallows the deletion of segments for the sake of completeness of the base. With respect to the positional relation between two strings, MAX constraints do not mention the directionality and locality. Further, the family of. 政治大. MAX constraints cannot rule out a possible candidate with the insertion since they are. 立. not concerned with the integrity of the truncated output. ANCHOR, like MAX, is a. ‧ 國. 學. type of faithfulness constraint. Furthermore, it imposes a positional correspondence. ‧. between two strings. Specifically, the truncated output must share constituents with. Nat. sit. al. er. io. 2.6 Summary. y. the base on the same edge.. n. v i n In brief, clipping and blendingC have surveyed inUrecent years. Many studies h ebeen ng chi. claim that they should be part of the grammar of language. Nevertheless, few researchers have provided a systematic account for the internal structure of the morphological processes from a phonological perspective. The use of the family of MAX constraints in Lappe’s study (2008) is not enough to account of the positional faithfulness between a blend and its sources. In addition, variations in clipping and blending have not been characterized in more detail. Following this discussion, I.

(31) 21. propose the application of Cophonology which makes it possible to analyze the variations. Chapter 3 and Chapter 4 will demonstrate how OT and Cophonology account for the internal structure of English clipping and blending, respectively.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v.


(33) CHAPTER 3. ENGLISH CLIPPING: AN OT ANALYSIS. This chapter examines the segmental preservation of English clipping from morphological and phonological perspectives. The analysis of the process is based on Correspondence Theory. The relation between the clipped output and its source is considered Output-to-Output correspondence. Under the truncation model (Benua 1995), the mapping relation between a clipped word and its source can be treated as follows:. 政治大. 立. ‧ 國. 學. (13) Clipping: laboratory lab SC-Identity (OO-FAITH). ‧. Base(Source): - - - - - - - - - - - - - - - - - - Clipping:. sit. y. Nat. io. n. al. []. er. [] IO-FAITH Input:. Ch. engchi. i n U. v. //. In the model of (13), Input-to-Output correspondence (IO-Faith) relates the input to the source, and Output-to-Output correspondence (OO-Faith) relates the source to the clipped form. The clipped output derives from the preservation of the segment of the source. The SC-identity is regulated by the constraints in the following discussion.. 23.

(34) 24. Chapter 3 is organized as follows: Section 3.1 introduces clipping types that preserve the morpheme of the source, and provides an OT account for the different types. Section 3.2 introduces clipping types that retain the phonological constituents of the source, and also takes an OT approach to account for the derivation of the different types. Section 3.3 presents ranking contradictions under classic OT, and employs a cophonological theory to solve inconsistencies. Section 3.4 summarizes this chapter.. 立. An OT Account for Morphological Clipping. 學. ‧ 國. 3.1. 政治大. This section addresses the clipped outputs which preserve the morpheme of the. ‧. source. It presents two types that correspond to different positions of the source and. sit. y. Nat. io. al. n. clipping.. er. demonstrates how ANCHOR constraints determine the two types of morphological. Ch. engchi. i n U. v. In English, there are two kinds of clipping: morphological and phonological clipping. Both of them show the correspondent constituents on one of edges of the source. Firstly, I divide morphologically clipped forms into two types in terms of the edge of morphemic preservation, as in (14) and (15)..

(35) 25. (14) Left edge examples →. bi-sexual porn-ography auto-mobile tele-vision. → → →. bi porn auto tele. (15) Right edge examples para-chute tele-phone. →. turn-pike news-paper. →. → →. chute phone pike paper. 政治大 The examples in (14) and (15) indicate that morphological clipping can be derived 立. ‧ 國. 學. from the left or right edge morphemic preservation of the source.. ‧. Positional faithfulness requires that two strings share the same constitutes at the. sit. y. Nat. designed periphery. McCarthy and Prince (1993a, 1995a,b) make use of ANCHOR. n. al. er. io. constraints in (7) to account for the positional faithfulness between the reduplicant. Ch. i n U. v. and its base. In this current research, I utilize ANCHOR constraints to analyze the. engchi. morphemic preservation of clipping. Relevant constraints are proposed in (16) to (18). (16) LEFT-ANCHOR-MORPHEME (Source, Clipping) Assign one violation mark for every first morpheme of the source that is not on the left edge of the clipped form..

(36) 26. (17) RIGHT-ANCHOR-MORPHEME (Source, Clipping) Assign one violation mark for every last morpheme of the source that is not on the right edge of the clipped form. (18) MorphemeMAX (abbreviated as MorMAX in tableaux) Assign one violation mark for every clipped output that has more than one morpheme.. 政治大. Constraint (18) poses a restriction on the length of the clipped output. It keeps the. 立. clipped form from having more than one morpheme. The interaction of the constraints. ‧ 國. 學. (16)-(18) is shown in (19). ‧. al. vision. *!. c.. television. *! (vision). d.. teleri. *! (ri). n. b.. Ch. engchi. RIGHT- ANCHORMORPHEME. sit. LEFT- ANCHORMORPHEME. er. tele. MorMAX. io.  a.. Nat. tele-vision. y. (19) {ANCHOR LEFT, MorMAX}>> ANCHOR RIGHT. i n U. *. v. *. First, tableau (19) compares the left-edge preserving candidate tele with the right-edge preserving candidate *vision, showing a conflict. Since the winner tele is left-anchored, it obeys ANCHOR LEFT but violates ANCHOR RIGHT. The loser *vision satisfies ANCHOR RIGHT at the expense of corresponding to the initial morpheme of the source. This is a violation of ANCHOR LEFT. For tele to be more optimal, ANCHOR LEFT must dominate ANCHOR RIGHT. Second, the candidates.

(37) 27. tele and *television indicate the relation of the ranking of MorMAX and ANCHOR RIGHT. The full form of candidate (c) incurs a violation of MorMAX because it contains more than one morpheme. For tele to be more optimal, MorMAX must dominate ANCHOR RIGHT. Finally, candidate *teleri is also ruled out by MorMAX. As a result, it is even less suitable than tele. Tableau (20) rearranges ANCHOR LEFT and ANCHOR RIGHT. Because the. 政治大. winning candidate phone is right-anchored, it obeys ANCHOR RIGHT but violates. 立. ANCHOR LEFT. The losing candidate *tele satisfies ANCHOR LEFT at the sacrifice. ‧ 國. 學. of corresponding to the final morpheme of the source. This incurs a violation of. ‧. ANCHOR RIGHT. For phone to be more optimal, ANCHOR RIGHT must outrank. Nat. io. sit. y. ANCHOR LEFT.. n. al. er. (20) ANCHOR RIGHT >> ANCHOR LEFT tele-phone a..  b. c.. Ch. MorMAX. engchi. tele. v. LEFT - ANCHORMORPHEME. *!. phone telephone. i n U. RIGHT - ANCHORMORPHEME. * *!. Nelson (2003) has challenged the assumption that right anchoring is equally important to the truncated forms which are faithful to the right edge of the source and has suggested that ANCHOR RIGHT is unnecessary. Making use of the idea of positional faithfulness, Nelson posits ANCHOR constraints only target prominent positions, as.

(38) 28. in (21). (21) Positional Anchoring (Nelson, 2003) a. Anchoring can target the initial position (important for root access). b. Anchoring can target a stressed position (acoustically prominent). c. The right edge does not qualify as a target for anchoring. According to Nelson’s proposal, (21a) explains why a large majority of English. 政治大. clippings are preserved from the left edge of the source and (21b) implies that there is. 立. (22) MAX-σ  (Base, Reduplicant) (Nelson 2003: p. 6). 學. ‧ 國. a constraint that requires the preservation of the stressed syllables, as (22) shows.. ‧. Each segment in the main stressed syllable of the base must have a correspondent. io. sit. y. Nat. in the reduplicant.. al. er. Instead of ANCHOR RIGHT, Nelson incorporates MAX-σ  to account for the. n. v i n C hforms and concludes right-edge preservation in the truncated e n g c h i U that some cases show right-anchored effects because of the stress on the final syllable. From Nelson’s point. of view, the actual preserving unit is the stressed syllables rather than the constituents on the right side, as (23) shows. (23) Nickname σ, ≠σ1 (Nelson, 1998) Amanda Virginia Elizabeth Rebecca.    . Mandy Ginnie Liz, Lizzy Becky.

(39) 29. However, left anchoring and MAX-σ  alone are not enough to account for the derivation of the clipping. As presented in example (15), the preserved units are always right-edge constituents but they are not stressed syllables such as chute from parachute or phone from telephone. In this thesis, I propose that the family of anchoring constraints consists of not only ANCHOR LEFT but also ANCHOR RIGHT. ANCHOR LEFT and ANCHOR RIGHT are re-ranked to account for the. 政治大. preservation of (14) and (15), respectively. When the last morpheme of the source has. 立. more than one syllable, ANCHOR LEFT may dominate ANCHOR RIGHT to select. ‧ 國. 學. the first morpheme as the clipped output, as in (14). When the last morpheme of the. ‧. Nat. 3.2 An OT Account for Phonological Clipping. al. er. io. sit. the last morpheme as the clipped output, as in (15).. y. source has one syllable, ANCHOR RIGHT may dominate ANCHOR LEFT to select. n. v i n This section addresses C theh clipped outputs which e n g c h i U preserve the phonological. constituent of the source. It presents three types that correspond to different positions of the source and shows that different types are explained by the interaction of the faithful and markedness constraints. In addition to morphological clipping, there are several other types of clipping that are phonologically preserved. In this section, I present some data that do not preserve a complete morpheme of the source, showing that the preserving unit is not.

(40) 30. necessarily a morphemic constituent. Phonological clipping, according to the edge of preservation, exhibits three patterns, being separated to the left or right or neither edge of the preservation of the source, as presented in (24)-(26). (24) Left edge examples moment laboratory. →. mo lab. →. delicatessen → examination →. deli exam. (25) Right edge examples →. 立. → →. ‧. ‧ 國. →. cause gator zine burb. 學. because alligator magazine suburb. 政治大. (26) Medial examples. io. y. sit. →. tec flu. n. al. er. →. Nat. detective influenza. Ch. i n U. v. Regarding (24) to (26), I make the following generalizations. First, since a clipping. engchi. derives from part of the source, the entire preservation is disallowed. Second, like morphological clipping, the stressed syllable can be truncated in phonological preservation. Thus, I suggest that the stress is not a key point in preservation of clipping. Third, since clipping prevents the preserving string from being deleted internally, the constituents are often retained contiguously. Finally, the size of a clipping is finite because we find that the structure of a majority of the clipped forms.

(41) 31. is either monosyllabic or disyllabic. Thus, the size of the clipping seems to be predictable. This section examines phonological variations of clipping through the use of constraint rankings. I propose the relevant constraints on the basis of the generalizations noted above. These constraints interact with one another to account for each type in the following discussion. 3.2.1. 政治大. Left-edge Preservation. 立. Given example (24) above, I further partition left-edge preserving clippings into. ‧ 國. 學. four types in terms of prosodic preservation. First, a left-edge preserved clipping may. ‧. preserve the first syllable such as mo from moment. Second, it may retain the first. Nat. io. sit. y. syllable and the second onset such as lab from laboratory. Third, it may preserve the. al. er. first two syllables such as deli from delicatessen. Finally, in addition to the first two. n. v i n C hclipping may keepUthe third onset such as exam from syllables, a left-edge preserved engchi examination. Three relevant faithfulness constraints on the four types of left-edge preserving clippings are defined in (27)-(29). MAX SC in (27) requires that each segment in the source has a correspondent in the clipping. In other words, it prevents the source from segmental deletion. CONTIGUITY in (28) stresses that the portion of the clipping in correspondence to the source forms a contiguous string and vice versa. In terms of.

(42) 32. CONTIGUITY, the internal deletion of the string of segments is banned. MAX-σ in (29) obligates the preservation of the stressed syllable of the source. (27) MAX (Source, Clipping) Assign one violation mark for every segment in the source that is not in the clipping. (28) CONTIGUITY (Source, Clipping) — ―No skipping, no intrusion‖. 政治大. Assign one violation mark for the string in the source that is deleted internally in. 立. 學. ‧ 國. the clipping.. (29) MAX-σ  (Source, Clipping). ‧. Assign one violation mark for every stressed syllable of the source that is not in. io. sit. y. Nat. the clipping.. er. The fact that the clipping is a contiguous sequence satisfies CONTIGUITY but that it. al. n. v i n C hviolates MAX SC.UThe stressed syllable is derives from a part of the source always engchi sometimes preserved but sometimes truncated. As (30) shows, I argue that CONTIGUITY dominates MAX-σ and MAX SC. (30) CONTIGUITY >> {MAX-σ, MAX SC} delicatessen. CONTIGUITY.  a.. deli. b.. dete. *!. c.. delict. *!.  d.. delicatessen. MAX-σ. MAX SC. * (te). catessen (8) li, tessen (8). * (te). a, essen (6).

(43) 33. Tableau (30) attests the ranking argument between CONTIGUITY and MAX. We compare the winning candidate deli with the losing candidate *dete. Since deli lacks the preservation of the stressed syllable, it violates MAX SC and MAX-σ. The losing candidate *dete obeys MAX-σ at the expense of forming a contiguous string. This is a violation of CONTIGUITY. For deli to be optimal, CONTIGUITY must govern MAX-σ. Candidate *delict violates all three constraints but it is more favored by. 政治大. MAX SC over the winner. In this case, for deli to be optimal, CONTIGUITY must. 立. dominate MAX to remove *delict. This tableau, however, cannot rule out the full. ‧ 國. 學. form of candidate (d) that respects CONTIGUITY, MAX-σ, and MAX SC. This. ‧. problem is solved in the following discussion. Since all clipped forms are defined. Nat. io. sit. y. with reference to the partial preservation of the source, it must be smaller than the full. n. al. er. form. In the following discussion, I will use a markedness constraint to account for their size.. Ch. engchi. i n U. v. ANCHOR constraints play an active role in the different types of clipping. Specifically, different anchorings interact with CONTIGUITY, MAX, and markedness constraints to generate various clipping types. The first type that retains the first syllable is evaluated from the interaction of CONTIGUITY, MAX SC, CWM(μμ), and LEFT-ANCHOR-σ as defined in (31) and (32)..

(44) 34. (31) ClippedWordMaximal (μμ) (abbreviated as CWM (μμ) in tableaux) Assign one violation mark for every clipped output that has more than two moras. (32) LEFT-ANCHOR -σ (Source, Clipping) Assign one violation mark for every first syllable in the source that is not the first syllable in the clipped output. The constraint in (31) requires that the size of a clipping is constrained by a bimoraic. 政治大. template and constraint (32) demands that the first syllable of the source has to be. 立. preserved to form the first syllable of the clipping. The ranking argument is presented. ‧. ‧ 國. 學. in (33).. (33) {CWM(μμ), LEFT-ANCHOR-σ} >> MAX SC. . . . al. n. . *!(μμμ). Ch. sit. y. MAX-σ. *. engchi U. v ni. MAX SC (4). er. . io.  a.. CONTIGUITY CWM(μμ) LEFTANCHOR -σ. Nat. moment . (3) *. (2). ([] is a long vowel that obtains two moras; all codas share one mora.) Tableau (33) elaborates that bimoraic and left-anchoring constraints rank above MAX SC. We compare the candidates [] with *[], finding that the losing *[] fatally violates CWM(μμ) because of being parsed into three moras. The losing *[] is more favored by MAX SC over the winning candidate. For the winning [] to be optimal, CWM(μμ) must dominate MAX SC. Consider the pair [] and.

(45) 35. *[], the losing *[] is also favored by MAX SC over []. Nevertheless, it fatally violates ANCHOR LEFT due to the right edge preserved syllable. To remove the loser, ANCHOR LEFT must dominate MAX SC. In (33), the losing candidate *[] may have two moraic structures. The first is that the long vowel obtains two moras and all coda consonants share the third mora, as in (34a). It is referred to as the superheavy moraic structure (Moren 2003). The. 政治大. second is the heavy moraic structure that the long vowel obtains two moras and shares. 立. the second mora with the coda consonant, as in (34b). I employ NOSHAREMORA to. ‧ 國. 學. rule out the heavy moraic structure, as given in (35) and (36).. ‧. (34) adapted from Moren (2003). μ. n. al. μ. m. o. m. Ch. m. engchi. i n U o. y. sit. μ. io. μ. b. Heavy. μ. er. Nat. a. Superheavy. vm. (35) NOSHAREDMORA (Broselow et al. 1997): Moras may not be shared. (36) moment .  a.. NOSHAREDMORA. CWM(μμ). . . superheavy. .  (heavy). *!(μμμ) *!(μμ). In (36), candidate (c) is ruled out because the constraint NOSHAREDMORA favors (a) over (c). Candidate (b) is removed by CWM(μμ). In this study, NOSHAREDMORA is.

(46) 36. ranked higher than other constraints on the moraic structure. Thus, the following discussion is talking about the candidates that always respect NOSHAREDMORA. All tableaux exclude. NOSHAREDMORA since the constraint is irrelevant to select the optimal output. No clipped forms change the feature of segment of the source. In other words, they always respect IDENT [F] that prevents the featural change of the source, as given in (37) where the irrelevant constraints in (33) are omitted.. 政治大 CWM(μμ) L -A. (37) ([] is a short vowel that obtains one mora; all codas share one mora.). .  . EFT. NCHOR-σ. 學.  a.. 立. IDENT [F]. *!. ‧. ‧ 國. moment . In the case of (37), candidate (b)IDENT [F] because it changes the. y. Nat. al. er. io. sit. feature of the vowel. Since candidate (a) respects all constraints, it is more harmonic. n. than (b) and is thus selected as the optimal output.. Ch. engchi. i n U. v. Concerning ANCHOR LEFT and MAX-σ, the ranking argument is attested by another example of the first type, as shown in (38). (38) LEFT-ANCHOR -σ >> MAX-σ CONTIGUITY CWM(μμ) LEFTgymnasium  ANCHOR –σ  a.  . . *!. MAX-σ *. MAX SC (6) (9). Tableau (38) compares the left-edge preserved candidate [] with the right-edge preserved candidate *[]. The left-anchoring candidate [] violates MAX-σ since.

(47) 37. it does not preserve the stressed syllable of the source. Although the losing *[] satisfies MAX-σ, it violates ANCHOR LEFT because it is not left-preserved. To select [] as the optimal output, ANCHOR LEFT must outrank MAX-σ. MAX-σ and MAX SC are in a stringency relation2. MAX SC is more stringent than MAX-σ if every violation of MAX-σ is also violation of MAX SC. Their ranking is not directly argued but it is inferred from the domain of constraint domination. That is, the. 政治大. specific constraint MAX-σ is placed above the general constraint MAX SC.. 立. The ranking shown in (38) is blocked in the evaluation of ‘streptococcus’ when. ‧ 國. 學. selecting the undesired candidate, as provided in (39).. CONTIGUITY CWM(μμ) LEFTANCHOR –σ. . . . sit. . n. al. er. a.. io. . y. Nat. streptococcus . ‧. (39). Ch. engchi. i n U. v. MAX-σ. MAX SC. *. (7)!. *. (6). In the ranking of (39), the loser *[] favored by MAX SC would be selected as the winner. I use the *COMPCODA constraint to solve the blocking, as defined in (40). (40) *COMPLEX-CODA (abbreviated as *COMPCODA in tableaux) Assign one violation mark for every cluster in coda position.. 2. McCarthy (2008) proposes that two constraints are in a stringency relation if every violation of one constraint is also a violation of the other..

(48) 38. Constraint (40) keeps coda position of clipping from having more than one consonant. Its effect is illustrated in (41). (41) *COMPCODA >> MAX SC MAX-σ *COMPCODA MAX SC CONT CWM(μμ) LEFTstreptococcus  ANCHOR –σ (7) *  a.  . . *! (). *. (6). 政治大. In (41), MAX SC favors the loser *[] over the winner []. The losing. 立. *[] causes a fatal violation of *COMPCODA since it has more than one. ‧ 國. 學. segment in coda. Meanwhile, the winner [] satisfies *COMPCODA. To select. ‧. [] as the optimal output, *COMPCODA must command MAX SC.. y. Nat. MAX-onset is proposed for this pattern, as in (42).. n. al. Ch. (42) MAX-onset (Source, Clipping). engchi. er. io. sit. In addition to the first syllable, the second type preserves the second onset.. i n U. v. Assign one violation mark for every onset in the source that is not in the clipping. Constraint (42) maximizes the preservation of all of the onsets of the source. In other words, it requests that every onset in the source has a correspondent in the clipped output, as evaluated in (43)..

(49) 39. (43) CONTIGUITY >> MAX -onset SC laboratory . . a. . CWM (μμ). CONT. LEFTMAX MAXANCHOR-σ -onset σ. *COMP CODA. .   *! (). MAX SC  (6) (4). . In the case of (43), I argue that CONTIGUITY dominates MAX-onset. Candidate *[skips []-preservation. This is a violation of CONTIGUITY. On the other. 政治大 hand, it is more favored by MAX-onset over the winner. To rule out the loser, 立. ‧ 國. 學. CONTIGUITY has to control MAX-onset.. ‧. The third type retains the first two syllables of the source. The constraint ranking. sit. y. Nat. in (43) is repeated to explain how this pattern generates, as in (44).. . a. . al. n. delicatessen []. CONT. CWM (μμ). Ch. LEFTANCHOR -σ. engchi.  . er. io. (44) The evaluation for the third type. i n U. v. MAX -onset. MAXσ. *COMP CODA. MAX SC. . . . (7)  (8)!. . . *!(μμμ). .  (6). In (44), candidate *[ is ruled out by the higer-ranking CWM(μμ) due to the parsing of the trimoraic structure. Candidate *[] is excluded by MAX SC because it preserves fewer constituents than []. As a result, candidate [] emerges as the.

(50) 40. optimal output. The constraint ranking in (44) cannot account for the fourth type where the first two syllables and the third onset are preserved, as (45) presents. (45) examination . CONT. CWM (μμ). LEFTANCHOR. MAX -onset. MAXσ. *COMP CODA. MAX SC. –σ. .  . . *. . 政治  大. *. (6) . . *. *! (4μ). . 立*! (3μ).  . . *! (5μ). *. (8). 學. . . ‧ 國.  a.. *. . (7) . ‧. In the evaluation of tableau (45), candidates [], *[], and *[]. y. Nat. er. io. sit. violate CWM(μμ), but *[] does not. Thus, the ranking predicts incorrectly that the. al. v. n. loser *[] is optimal. To solve the blocking, I apply LEFT-ANCHOR-Σ to the. Ch. engchi. i n U. operation of the fourth type of clipping, as defined in (46). (46) LEFT-ANCHOR-Σ (Source, Clipping) (Σ=σσ3). Assign one violation mark for every first foot of the source that is not the first foot of the clipped output. Constraint (46) requests that the first and second syllables of the source must be the first and second syllables of the clipped output. Its function is shown in (44), where. 3. McCarthy and Prince (1995a) have proposed that Feet are binary syllabic or moraic..  (4).

(51) 41. CONTIGUITY is irrelevant to evaluation and omitted. (47) examination . CWM (μμ). . * (4μ). a.. . LEFTANCHOR -Σ. LEFTANCHOR -σ. MAX MAX-onset σ . *COMP CODA. MAX SC . *. (6) . . . *. *. *. .  (8). 政治大 construction while the loser *[] violates L -A 立. In (47), the winner [] violates CWM(μμ) because of the quadramoraic NCHOR-Σ. EFT. because it fails to retain. ‧ 國. 學. the second syllable completely. As the ranking shows, CWM(μμ) and LEFT-ANCHOR. sit. Nat. favors [] over *[], the optimal output is [].. y. ‧. -Σ are unranked. Consequently, MAX-onset is active for the evaluation. Since it. n. al. er. io. The evaluation of another candidate pair does not yield to the ranking of (47). Ch. i n U. v. even though LEFT-ANCHOR-Σ is adopted, as in (48). Constraints CONTIGUITY and. engchi. LEFT-ANCHOR-σ are undominated. Thus, the tableau (48) discusses the candidates that obey CONTIGUITY and LEFT- ANCHOR-σ and excludes the undominated constraints. (48) examination . CWM (μμ).  a.  . *(4μ). .  *(5μ). LEFTANCHOR -Σ. MAX -onset. MAXσ. *COMP CODA. MAX SC. . *. (6). . *. (4). Tableau (48) indicates that LEFT- ANCHOR-Σ favors neither the winner nor the loser.

(52) 42. and both of them incur a violation of CWM(μμ). Again, MAX-onset serves the function of removing the actual output []. The ranking needs to be modified. It is clear that the candidate pair in (48) is not restricted to a bimoraic structure. They are confined to the disyllabic template. Accordingly, I propose CWM (σσ) to operate the type, as defined in (49). (49) ClippedWordMaximal (σσ) (abbreviated as CWM (σσ) in tableaux). 政治大. Assign one violation mark for every clipped output that has more than two. 立. syllables.. ‧ 國. 學. Constraint (49) targets the size of the fourth type, specifying that a clipping must be a. ‧. disyllabic form. The constraint ranking is argued in (50). Constraints CONTIGUITY. Nat. io. sit. y. and LEFT-ANCHOR are undominated and thus omitted in the tableau.. n. al. er. (50) CWM(σσ) >> MAX –onset SC examination . CWM (μμ).  a.. * (4μ). . .  *(5μ). Ch. CWM (σσ) *!. MAX -onset . engchi . i n U. MAX-σ. v. *COMP CODA. MAX SC. *. (6). *. (4). As Tableau (50) presents, CWM(σσ) dominates MAX -onset. Since the winner [CWM(σσ).The loser *[] is more favored by MAX-onset at the cost of retaining a trisyllabic structure. This incurs a violation of CWM(σσ). To eliminate *[], CWM(σσ) must dominate MAX-onset. Now, the ranking arguments for the left-edge preserved clippings are established.

(53) 43. in (51). (51) Ranking arguments for four types of left-edge preserved clipping a.. CONTIGUITY >> MAX-σ >> MAX SC. b.. { CWM(μμ), LEFT-ANCHOR-σ} >> MAX SC. c.. LEFT-ANCHOR-σ >> MAX-σ. d.. *COMPCODA >> MAX SC. e.. CONTIGUITY >> MAX-onset. f.. CWM(σσ) >> MAX-onset. 立. 政治大. ‧ 國. 學. As (51) represents, the least dominant constraint MAX SC is directly governed by. ‧. MAX-onset and MAX -σ, which are unranked with respect to each other. The. Nat. io. sit. y. constraint *COMPCODA ranks above MAX-onset and MAX –σ; however, it is. er. dominated by a set of unranked constraints which consist of CONTIGUITY,. al. n. v i n CWM(μμ), CWM(σσ), LEFTC -Ah NCHOR-σ, and LEFTU e n g c h i -ANCHOR-Σ. The ranking is summarized as follows: (52) Summary ranking for phonological clipping (left-edge) { CONTIGUITY, CWM(μμ), CWM(σσ), LEFT-ANCHOR-σ, LEFT-ANCHOR-Σ }. *COMPCODA {MAX –onset, MAX-σ} MAX SC.

(54) 44. Summary tableaux (53)-(56) elaborate the analyses of the left-edge preserving clipping, showing the process which leads to the production of the first type in (53), the second in (54), the third in (55) and the fourth in (56). (53) The first type moment . CON. CWM (σσ). CWM (μμ). LEFTANCHOR. *COMP CODA. MAX MAX -σ -onset. MAX SC. -σ. . (4) (3). . . . (2). *COMP MAX MAX -σ CODA -onset. MAX SC. ‧ 國. *. Nat. . CWM (μμ). LEFTANCHOR -σ  . *. *. *. ‧. . (σσ). RIGHT. 學. streptococcus CON . 立 CWM. 政治大 *!(3μ) * *. *! (pt). *. y. . . . . io. n. al. . v. In summary tableau (53), it is further argued that *COMPCODA dominates. Ch. engchi. i n U. MAX –onset in the evaluation of ‘streptococcus’. MAX-onset favors the losing candidate *[] over the winning [], since *[] with a non-falling coda violates the constraint *COMPCODA, which is obeyed by []. For the winning [] to be optimal, the MAX-onset must rank below *COMPCODA. Again, another constraint ranking is argued, as in summary tableau (54)..  (7). sit. .  a.. . . er.  a..  (6).

(55) 45. (54) The second type laboratory .  a.. CON. CWM (σσ). . CWM (μμ). LEFTANCHOR -Σ. *. *COMP CODA. MAXσ. *. MAX -onset. MAX SC. . . . (6) .  .  *! (). . *. (4). Summary tableau (54) offers information of the ranking relation between. 政治大 and thus fulfills the CONTIGUITY constraint. Nevertheless, it violates CWM(σσ) due 立. CONTIGUITY and CWM(σσ). The winning candidate [is a contiguous string. ‧ 國. 學. to its monosyllabic form. The losing candidate *[] obeys CWM(σσ) but violates. ‧. CONTIGUITY. To select the winning candidate as an optimal output, CONTIGUITY. sit. y. Nat. must dominate CWM(σσ).. . a.. al. n. delicatessen []. er. io. (55) The third type. CON. CWM (σσ). Ch. CWM (μμ). engchi. . i n U. LEFTANCHOR -Σ . v *COMP CODA. MAX MAX -σ -onset . MAX SC  (7). . . *!. *. . .  (8). . . *!(3μ). . . .  (6).

(56) 46. (56) The fourth type examination .  a.. CON. CWM (σσ). . CWM LEFT(μμ) ANCHOR -Σ  *. *COMP CODA. MAX MAX -σ -onset . . . *!. . *. * (5μ).  (6). (4μ) . MAX SC. . . *. *. . . . (4)  . 治政大a ranking argument that In evaluating the candidate pair [] and *[], we find 立 ‧ 國. 學. places CWM(σσ) above CWM(μμ). The point of tableau (53) is to explain why a clipping may be confined to a disyllabic rather than bimoraic template. Since the. ‧. higher-ranking disyllabic constraint prefers [] to *[], candidate [] is. n. al. er. io. sit. y. Nat. optimal.. 3.2.2 Right-edge Preservation. Ch. engchi. i n U. v. We have now looked at a number of English clipped forms that are left-anchored under an OT explanation. Nevertheless, we still need to outline the derivation of a few right-anchored data. In this section, I provide the right-edge preserving examples with an account of ANCHOR constraints under the framework of Optimality Theory. See examples in (574):. (8).

(57) 47. (57) Right edge examples because alligator magazine suburb. →. cause gator zine burb. → → →. Drawing an analogy to left-anchored clipped outputs, the size of the preserving portion in (57) can be either monosyllabic or disyllabic constituent. Likewise, the stressed syllable is not always retained. As for right-edge preserving types, I propose. 政治大. the use of ANCHOR RIGHT instead of ANCHOR LEFT. The first type preserves the. 立. final syllable of the source and thus I suggest RIGHT-ANCHOR-σ should be active in. ‧ 國. 學. the summary ranking of English clipping, as defined in (58).. ‧. (58) RIGHT-ANCHOR-σ (Source, Clipping). sit. y. Nat. io. n. al. final syllable in the clipped output.. Ch. engchi. er. Assign one violation mark for every final syllable in the source that is not the. i n U. v. Constraint (58) requires that every final syllable of the source has a correspondent on the right edge of the clipping. It interacts with other constraints, as operated in tableau (59)..

(58) 48. (59) RIGHT -ANCHOR-σ >> *COMPCODA >> MAX-σ suburb ..  a.. CON. CWM CWM RIGHT(σσ) (μμ) ANCHOR -σ  *. . *COMP MAXσ CODA. b.. . *. *!. * . . . *. *!. . * *. MAX -onset. MAX SC. .  (2). .  (3). .  (3). Tableau (59) verifies that RIGHT-ANCHOR-σ dominates *COMPCODA and MAX-σ. We compare the right-edge preserving candidate [] with the left-edge preserving. 政治大. candidate *[]. At the cost of preserving the final syllable, candidate *[]. 立. satisfies *COMPCODA and MAX-σ but violates ANCHOR RIGHT. Candidate []. ‧ 國. 學. respects ANCHOR RIGHT. To knock *[] out of the competition, ANCHOR. ‧. RIGHT must rank above *COMPCODA and MAX-σ. The losing candidate *[] is. y. Nat. n. al. er. io the optimal output.. sit. harmonically bounded4 and then removed. As a result, the winner [] emerges as. Ch. engchi. i n U. v. It is clear, however, that the ranking (59) fails to evaluate either the winner or the other disyllabic candidate, as (60) shows. (60) The blocking evaluation suburb ..  a.  b. 4.  . CON. CWM (σσ) *! . CWM (μμ). *(3μ). RIGHTANCHOR -σ . *COMP CODA. MAXσ. MAX -onset. MAX SC. *. *. .  (2). .  (1). *. According to McCarthy (2008), a harmonically bounded candidate cannot win no matter how the constraints are ranked..

(59) 49. As (60) presents, CWM(σσ) is active to eliminate the actual output. Since CWM(μμ) favors the winner over the loser, I suggest the ranking between CWM(σσ) and CWM(μμ) should be reversed in the evaluation of the first type, as modified in (61). (61) CWM(μμ) >> CWM(σσ) CON. suburb .. CWM (μμ). CWM (σσ). RIGHTANCHOR-. *COMP CODA. MAXσ. MAXonset. MAX SC. *. *. .  (2). .  (1). σ.  a. b.. . . *. . * 治政大 In (61), the ranking argument indicates CWM(μμ) ranks over CWM(σσ). The 立 *!(3μ). ‧ 國. 學. higher-ranking CWM(σσ) excludes the loser *[] successfully because the size. ‧. of the candidate is larger than two moras, leading to a fatal violation. The winner. sit. y. Nat. [] is selected as the optimal output.. n. al. er. io. The second right-edge preserving type retains the last two syllables of the source.. Ch. i n U. v. The ranking in (60) is used to account for its operation, as given in (62).. engchi. (62) CWM(σσ) >> CWM(μμ) alligator . CON. CWM (σσ). CWM (μμ). RIGHTANCHOR. *COMP CODA. MAX MAX -σ -onset. MAX SC. -σ.  a.  c..   . * *! *!. *. . *. . (3). . *. . . . . . Compared with (61), tableau (62) leads to an opposite ranking argument with regard to CWM(σσ) and CWM(μμ). It presents that CWM(σσ) dominates CWM(μμ) in the. (5)  (1).