Conversion as Zero derivation: Evidence from affix concatenation PUFS Yongsung Lee (yslee@pufs.ac.kr) 1) What is conversion? a. Conversion is the process by which lexical items change (grammatical) category without any concomitant change in form. (Lieber 2005: 418) b. Conversion is one the processes that may take part in the creation of new lexemes in English. (Valera 2005: 20) 2) Types of conversion a. Total conversion (Quirk et al. 1985) The converted words participate in morphological processes (derivation and inflection). bottle N bottle V bottler N, bottled V-PAST b. partial conversion (Quirk et al. 1985) The converted form does not show any derivation or inflection best A best N *bests N-PL, *bestable A c. approximate conversion (Kiparsky 1982) There is slight phonological difference between two pairs expórt V éxport N hou[s]e N hou[z]e V? sing V song N??breathe N breath V 3) Examples of conversion (Quirk et al. 1972: 1011-1013) a. Verb Noun desire, dismay, cover, retreat, break, turn, rise, catch, swim, look b. Adjective Noun daily, comic, valuable, constant, final (Alternative: Deletion of Nouns in Adj-N sequence.) c. Noun Verb forest, milk, tape, fish, cash, champion, experiment, mother d. Adjective Verb dry, empty, narrow, calm, humble, lower, idle 4) Different views on conversion a. Two distinct lexical entries (Homophony approach) (cf. Quirk et al 1985:70-71, cf. Don 2004) i. Two forms are homophonous. ii. Fails to capture the visible phonological, syntactic and semantic relations between the two. (cf. Kiparsky 1997, Don 2004:932) iii. Some words like long A and to long (to yearn) or bare A and bear N should be treated as homonyms. (cf. Jovanovič 2003: 426) b. Dual membership (Polysemy approach) (cf. Sanders 1988) i. Predecessor of the underspecification approach in (4e). NKAELL Fall Conference (Pusan University of Foreign Studies, Oct. 11, 2008) Page-1
ii. iii. A lexical item may have more than one grammatical category. linguists have generally tended to view one of the functions as more basic than the other. (Sanders 1988:155) fish (plural) fished (past) Nouns are irregular drinks (plural) drank (past)- Verbs are irregular c. Zero-derivation (Jesperson 1942, Marchand 1969, Don 2004, 2005) i. One is derived from the other. (Kiparsky 1982) ii. There are conspicuous common characteristics between conversion and derivation. iii. No specific argument for zero-derivation in the literature (Don 1993: 35) iv. How many zero-affixes are there and what are their semantic value? d. Empty category movement (Functional shift, relisting) (Williams 1981, Lieber 1981, Bauer 1983, Valera 2005) i. Is the zero affix a prefix or a suffix? ii. There is no derivation. A simple category-changing rule does the work. iii. This explains the regular nature of irregular verbs (grandstanded) (cf. baby-sitted, *baby-sat) iv. Conversion is not a process of creating a new word (but a new function). e. Underspecification (Hale and Keyser 1993, Marantz 1997, Barner and Bale 2002) i. Certain lexical categories are not marked in the lexicon. (Syntax fills in the category) ii. Unable to accommodate approximate conversion. (Kiparsky 1982), unless the stress assignment is assumed to take place after syntax not in the lexicon. iii. Unable to explain regular plurals but irregular past forms. (cf. (diii)) 5) Superiority of zero-derivation hyopothesis: a. Simplicity: No specific morphological status for conversion. Conversion, unlike derivation, may not be a universal morphological process. b. Overt relationship between conversion and affixation i. Argument structure of converted item. They destroy the city their destruction of the city Tom reviews the book Tom s review of the book cf. John grows tomatoes *John s growth of tomatoes John tapes the box *John s tape of the box ii. Directionality of conversion The converted verbs, just like derived verbs are generally transitive ones. N Adv, V Adv, Adv A, Adv N, Adv V mappings are not possible both in derivation and conversion iii. Synonymy blocking Blocking is an economy principle that can be thought of informally as an injunction to avoid coining synonyms: If you already have a perfectly good expression for something, don't invent another one. (Aronoff and Fuderman 2005:219) NKAELL Fall Conference (Pusan University of Foreign Studies, Oct. 11, 2008) Page-2
*to alchohol (to alchoholize) *to letter (to write) *to singer (to sing) Suffixes derive sbs(nouns) from vbs, and it would be contrary to reason to form such verbs as arrival, guidance, improvement, organization when arrive, guide, improve, organize exist. (Marchand 1969:302) iv. Relationship with phonological rules Zero-derivation analysis offers more coherent explanation for approximate conversions. [export] V [export] N (cf. [órigin]n [original]a) [hou[s]e] N [hou[z]e] V (cf. [electri[k]] A [electri[s]ity] N ) 6) Problems of zero-derivation hypothesis. a. The affixal nature of the zero-morpheme. (Is the zero affix a prefix or suffix?) (Strauss 1982) i. This, however, is a pseudo-problem given the Right-hand Headed Rule: Right-hand Head Rule (Williams 1982:248) In morphology, we define the head of a morphologically complex word to be the right-hand member of that word. ii. The head changes the part of speech, meaning that the conversion is controlled by adding a suffix, not a prefix in English. (with exceptions for such prefixes as {be-}, {de-} and {en-} b. The semantic range of the zero-morpheme. (what are the semantic characteristics of zero-affixes?) i. Examples from Marchand (1969: 300) bed spread a bed put to a bed go to bed sleep with ii. We have similar problems in defining the meaning of any overt suffix. c. Affix interaction (Does the zero-suffix belong to Class One or Class Two?) i. If it is Class one, then how come it is attached after class II affixes? valuables, engineer, waitress, ii. If it is Class Two suffix, why it is not attached to the majority of Class II suffixes? * These simply reflect the problem of Lexical morphology/phonology model. d. Derived word blocking i. Booij (2002: 136) It is not easy to find derivationally complex nouns that feed conversion.. ii. Don (2005:6) Also in English the correct generalization seems to be that affixed forms are not available for (verb-forming) conversion. iii. Jensen (1990: 95) NKAELL Fall Conference (Pusan University of Foreign Studies, Oct. 11, 2008) Page-3
It follows that nouns formed with level 1 suffixes can give zero-derived verbs,but not nouns formed with level 2 suffixes. * These observations are generally wrong! 7) Examples of derived word conversion ((Marchand 1969 and others) a. N V vacation audition auctioneer engineer patronage bandage package frontier barrier pressure adventure posture closure miniature waitress disrupture b. V N Review rematch preview restructure recapture c. A V bloody dirty weary negative impersonate opinionate alienate (maybe we can posit two different {-ate} suffix.)--designate d. A N valuable knowable printable harmonic phobic scientific Spanish Korean Chinese affirmative negative relative protestant adherent aspirant personal natural national collateral fundamental cautionary 8) Problems of Lexical morphology. a. Undergeneration (Bracketting paradox) Class One affixes are found after Class Two affixation [un-grammatical]-ity [govern-ment]-al b. Overgeneration Some possible affix combinations are not found. i. Class One-Class One combination *[institute-ion]-(i)an *[present-ation]-(i)an *[realize-ation]-ous *[personal-ity]-ous *[real-ity]-ize *[resident-y]-ize ii. Class One-Class Two combination *[real-ize]-ment *[sign-ify]-ment *[deform-ity]-ful *[personal-ity]-ful *[institute-ion]-ism *[antique-ity]-hood *[relate-ion]-ly *[perceive-tion]-ly iii. Class Two-Class Two combination *[kind-ness]-ful *[encourage-ment]-ful NKAELL Fall Conference (Pusan University of Foreign Studies, Oct. 11, 2008) Page-4
*[defend-ant]-ly *[reside-ent]-age *[develop-ment]-ly *[assemble-y]-age 9) Affix types (An alternative to Lexical morphology approach) a. Beginning suffixes (Fabb, 1988) (28 pout of 43 suffixes) * Some suffixes are added only to unsuffixed words. V-age (steerage) V-al (betrayal) V-ance (annoyance) V-ment (containment) V-y (assembly) N-age (orphanage) N-hood (nationhood) N-ism (despotism) N-an (librarian) N-ist (methodist) V-ant (defendant) V-ful (forgetful) V-ant (defiant) V-ory (advisory) V-ive (restrictive) N-ful (peaceful) N-ous (spacious) N-y (hearty) N-ly (ghostly) N-ish (boyish) N-an (reptilian) N-ed (moneyed) A-ly (harshly)** N-ate (originate) N-ify (classify) N-ize (symbolize) A-ify (intensify) * Underlined affixes are classified as Class II suffixes ** This might not be true. (luckily, alternatively ) b. Closing suffixes (Aronoff and Fuhrhop, 2002) * Some suffixes do not allow any other suffix after them. N-hood (neighborhood) N-ful] N (auditorium-ful) V-ful (regretful) N-ful] V (sorrowful) A-ity (reality) N-dom (kingdom) N-ly (lovely) N-less (careless) N-ship (scholarship) N-ling (duckling) N-y (speedy) V-ory (advisory) A-ly (happily) A-ness (kindness) * Underlined affixes are classified as Class I affixes. * {-ly} and {-ness} might be treated as appendix, meaning it can be attached even after closing suffixes. c. Monosuffix constraints and free suffixes (Aronoff and Fuhrhop, 2002) i. Many Germanic suffixes obey monosuffix constraint. Monosuffix Constraint Suffixes that select Germanic bases select unsuffixed bases. ii. This means that Germanic suffixes are the beginning and closing suffixes at the same time. d. Affix affinity (Plag 1999) Certain stem final suffixes have a strong affinity for certain predetermined suffixes. i. Stem final verb-forming affix {-ate} takes {-ion} for deverbal nouns. designate-ion impersonate-ion invigorate-ion alienate-ion ii. Stem final verb-forming affix {-ize} takes {-ation} to form nouns. realize-ation personalize-ation symbolize-ation materialize-ation iii. Stem final verb-forming suffix {-ify} takes {-cation} to form nouns. notify-cation signify-cation NKAELL Fall Conference (Pusan University of Foreign Studies, Oct. 11, 2008) Page-5
humanify-cation diversify-cation iv.stem final noun forming suffix {-ion} takes {-al} to form adjectives. relation-al correction-al location-al repetition-al 10) Relationship between conversion and affix concatenation In general, conversion of derived words is rare. Such rarity (and presence) of derived word conversion can only be explained with reference to affix types, if conversion is viewed as a process of derivation by adding zero morphemes. a. Conversion and beginning suffixes. i. In the framework of conversion as zero derivation, we can see that once conversion is applied, No beginning suffixes (nor monosuffixes) can be added after it. ii. This explains the rarity of further derivation from the converted words. iii. But still we can see that the closing suffix or free suffix may be added. (cf. total conversion in (2a).) milk N milk V milker N (*milk-ment N ) bottle N bottle V bottler N (*bottle-age A ) b. Conversion and closing suffixes. i. Once a closing suffix (or a monosuffix) is attached, the form cannot undergo conversion. ii. This explains the lack of conversion from many derived words. iii. But still it allows the conversion from derived words that have beginning or free suffixes. (see examples in (7a), (7c) and (7d).) c. Conversion and affix affinity i. Affix affinity wins over conversion in competition. symbolize [[symboliz] V -ation] N, *[[symbolize] V -Ø] N signify [[signify] V -cation] N, *[[signify] V -Ø] N alienate [[alienate] V -ion] N, *[[alienate] V -Ø] N ii. This explains the lack of conversion from a word that has a verb-forming suffix as in (7b). iii. And this also explains conversion from verbs, if the verbs do not have verb forming suffixes. (prefixes are OK!) 11) Conclusion a. The noticeable connection between conversion and derivation can only be expressed, when we view conversion as part of the derivation. b. The apparent problems of zero-derivation is related to the deficiency of Lexical Morphology. c. With newly proposed affix types, the former problems can be dealt with. d. The remaining issues are basically semantic, not morphological nor phonological NKAELL Fall Conference (Pusan University of Foreign Studies, Oct. 11, 2008) Page-6
References Aronoff, Mark and Kirsten Fuderman 2005. What is Morphology. Blackwell Publishing, Malden, MA. Aronoff, Mark and Nanna Fuhrhop. 2002. Restricting suffix combinations in German and English: Closing suffixes and the monosuffix constraints. Natural Language & Linguistic Theory 20: 451 490, 2002. Booij, Geert. 2002. Prosodic restrictions on affixation in Dutch. In Geert Booij and Jaap van Marle (eds.) Yearbook of Morphology 2001, 183-201. Kluwer Academic Publishers, Dordrecht/Boston/London. Barner, David and Allen Bale. 2002. No nouns, no verbs: psycholinguistic arguments in favor of lexical underspecification. Lingua 112, 771 791. Don, Jan. 1993. Morphological Conversion. Doctoral dissertation, Utrecht University. Don, Jan. 2004. Categories in the lexicon. Linguistics 42.5: 931 956. Don, Jan. 2005. On Conversion, Relisting and Zero-derivation: A comment on Rochelle Lieber: English word-formation processes. SKASE Journal of Theoretical Linguistics 2.2: 2-16. Fabb, Nigel 1988. English Suffixation is Constrained only by Selectional Restrictions, Natural Language and Linguistic Theory, 6.4: 527-539. Hale, Ken and Keyser, Jay. 1993. On the argument structure and the lexical expression of syntactic relations. In Hale, K. and J. Keyser (eds.), The View from Building 20, Essays in Linguistics in Honor of Sylvain Bromberger, 53 109. MIT Press, Cambridge, MA. Jensen, John T. 1990. Morphology: Word structure in Generative Grammar. John Benjamin Publishing Co., Amsterdam. Jespersen, Otto. 1942. A Modern English Grammar on Historical Principles. Part VI: Morphology. George Allen & Unwin Ltd., London. [reprinted in 1974] Jovanović, Vladimir Ž. 2003. On Productivity, Creativity and Restrictions on Word Conversion in English. Facta Universitatis Series in Linguistics and Literature 2.10: 425-436. Katamba, Francis 1993. Morphology. Palgrave, New York. Kiparsky, Paul. 1982. From Cyclic Phonology to Lexical Phonology. In van der Hulst, H and N. Smith (eds.), The Structure of Phonological Representations, 131-175. Foris Publications, Dordrecht. Kiparsky, Paul. 1997. Remarks on denominal verbs. In Alsina, Alex J. Bresnan and P. Sells (eds.), Argument Structure, 473 499. Stanford: CSLI. Lieber, Rochelle. 1981. On the Organization of the Lexicon. PhD diss. University of New Hamsphire. [reproduced by Indiana University Linguistics Club, Bloomington, IN.] Lieber, Rochelle. 2005. English Word-Formation Processes: Observations, Issues, and Thoughts on Future Research. In Štekauer, Pavol and Rochelle Lieber (eds.), Handbook of English Word-formation. Dordrecht: Springer, 375-427. Marantz, Alec. 1997. No escape from syntax: don t try morphological analysis in the privacy of your own lexicon. In Dimitriadis, A. et al. (eds.), Proceedings of the 21st Annual Penn Linguistics Colloquium: Penn Working Papers in Linguistics 4, 201 225. Marchand, Hans (1969). The Categories and Types of Present-Day English Word-Formation. A Synchronic-Diachronic Approach, second edition. C. H. Beck sche Verlagsbuchhandlung, München. Quirk, Randolph, Greenbaum, Sidney, Leech, Geoffrey and Svartvik, Jan. 1972. A Grammar of Contemporary English. Longman, London Quirk, Randolph, Greenbaum, Sidney, Leech, Geoffrey and Svartvik, Jan. 1985. A Comprehensive Grammar of the English Language. Longman, London. Plag, Ingo. 1999. Morphological Productivity: Structural Constraints in English Derivation. Topics in English Linguistics 28. Mouton de Gruyter, Berlin. Sanders, Genrald. 1988. Zero derivation and the Overt Analogue Criterion. In Hammond, Michael and Michael Noonan (eds.) Theoretical Morphology, 155-175. Academic Press, New York. Strauss, Steven. 1982. On relatedness paradoxes and related paradoxes. Linguistic Inquiry 13, 694-700. Valera, Salvador 2005. Conversion vs. unmarked word-class change. SKASE Journal of Theoretical Linguistics 2.1:20-42. Williams, Edwin. 1981. Argument structure and morphology. The Linguistic Review 1: 81-114. NKAELL Fall Conference (Pusan University of Foreign Studies, Oct. 11, 2008) Page-7
영접사파생으로본품사전환 - 접사화과정을중심으로 Conversion as Zero derivation: Evidence from affix concatenation 부산외대이용성 (yslee@pufs.ac.kr) 품사전환이란하나의어형태가모습은바뀌지않고다른품사로사용되는것을말한다. 이러한품사전환이 기능전이 등의고유의어형성과정인지아니면영접사를첨가하여만들어진일종의파생현상인지에대한논의가활발하다. 물론이외에도본디하나이상의품사를가지고있는것으로규정하는경우도있고또품사가줄여표기된 (underspecification) 것이라는주장도존재한다. 전통적으로는 Jesperson (1942) 이후품사전환을영접사에의한파생이라설명하였으나 (Marchand 1969) 이러한입장에대한대안이나타나고있는실정이다. 본고는이러한여러가지대안이있기는하지만여전히영접사파생으로보는시각이옳다는것을접사화과정을중심으로논의하고자한다. 먼저품사전환을영접사파생으로보는관점이지니는문제를검토하여이러한문제를해결할수있는방안을제시한다. 영접사가접두사인가접미사인가, 과연몇개의영접사를인정해야할것인가, 영접사의의미영역을어떻게규정할것인가등기존에제기된문제는사실상형태론의입장에서볼때에는그렇게큰문제가되지못할수도있다. 반면에품사전환이파생과밀접한관계가있다는것은영접사파생의방법이외에는설명할길이없다. 영접사파생은비파생어에서는매우광범위하게일어나지만, 파생어의경우는영접사파생이매우제한적이다. 그러나 Booij (2002) 의주장처럼파생어가결코영접사파생의대상이될수없는것은아니며 (to pressure, to vacation), Jensen (1990) 이설명하듯 2군접사가붙으면품사전환이일어나지않는다고할수없다. (a valuable, to engineer) 본논문은 (1) 품사전환이비파생어에서더흔하게일어나는이유가무엇인지와 (2) 어떤종류의파생어가품사전환을거치는지를접사화제약에연결하여설명하고자한다. Siegel이후에전통적으로받아들여진접사분류의 (1군접사, 2군접사 ) 문제점을제시하여이러한접사의분류가지니는괄호매김역설, 과소생성, 과다생성의문제를지적하고 Fabb (1988) 이후에연구되는접사의특성을재분류한다. 이에대한연구는진행중이기는하지만현재까지의연구로서는 (1) 일단첨가가되면그뒤에다른접사를허용하지않는접사 (closing affixes) 와 (2) 접사가이미첨가된단어에는붙지못하는접사 (beginning affixes) 에대한기존의연구를바탕으로이를영접사파생과연결시키고자한다. 먼저최초접사 (beginning) 와영접사파생을관찰한다. Fabb가조사한바에의하면그가연구한 43개의접미사중에서 28개가여기에속한다. 또한이러한접사에는기존의 1군접사 (V-al, V-y, N-ify) 와 2군접사 (V-ment, V-ful, N-ly, N-hood) 가섞여있음을알수있다. 품사전환이영접사파생이라면품사전환된단어에는최초접사가첨가되어단어를파생시킬수없을것이다. 사실상 arrival, signify, development등의단어는실제로품사전환으로다른품사가될수없다. 이러한공백은결코우연한것이아니며이를설명하기위해서는품사전환을파생으로인한접사화의방법으로보지않으면안될것이다. 또한 Aronoff & Fuhrhop (2002) 는영어의접사가운데에는자신의뒤에다른접사를인정하지않는최후접사 (closing affixes) 가있음을확인해내었다. 품사전환이일종의파생이라면파생을일으키는영접사는이러한최후접미사 (closing suffixes) 의뒤에는첨가될수없을것이다. {-ity} 의경우기존의
1군접사로분류하였지만 Aronoff & Fuhrhop는이를최후접미사로분류하였다. 따라서 personality와같은단어는마지막에 {-ity} 가첨가되어있으므로품사전환을통해새로운품사를얻을수없음을알수있다. 실질적으로 Aronoff & Fuhrhop의연구에따르면실질적으로상당수의 2군접사가최초접사이며동시에최후접사이다. 즉이러한접사는오직비파생어에한번붙음으로서파생이완료가되며이러한일체의접사가붙은단어는역시영접사파생에의한품사전환이일어나지않을것임을분명하게예측할수있다. 아울러최초접사도최후접사도아닌접사의경우는품사전환이자유롭게일어나서 vacation, bandage, auctioneer, pressure (N V), bloody, negative, alienate (A V), 그리고 valuable, hamonic, Spanish, affirmative, aspirant, personal (A N) 등의파생어가품사전환을일으킬수있음을체계적으로설명할수있다. 이러한부분은품사전환을영접사파생이아닌다른방법으로는설명할수없는것이다. 이같은관점에서볼때영접사파생에의한품사전환이라는가설은의미해석의문제가있음에도불구하고여전히가장뛰어난설명력을지닌방식임을알수있다.