“building words via metonymy: a comparison of russian, czech and norwegian” laura a. janda...

48
“Building Words via Metonymy: A Comparison of Russian, Czech and NorwegianLaura A. Janda Universitetet i Tromsø

Post on 21-Dec-2015

213 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

“Building Words via Metonymy: A Comparison of Russian,

Czech and Norwegian”

Laura A. Janda

Universitetet i Tromsø

Page 2: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

2

Main Idea• Role of metonymy in grammar

– Metonymy as the motivating force for word-formation

– Metonymy is more diverse in grammar than in lexicon

• Why this has been previously ignored– Most linguistic research on metonymy has

focused on• lexical phenomena• languages with relatively little word-

formation

Page 3: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

3

Overview1. The Big Picture: why study metonymy in grammar?• Cognitive structure of information

2. Relevant Previous Scholarship3. Databases: Russian, Czech, Norwegian• Size & structure of databases• Metonymy & Word class designations• Specificity of suffixes

4. Observations• Comparison across domains (lexicon vs. grammar)• Directionality of metonymy• Comparison across languages

5. Conclusions

Page 4: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

4

1. The Big Picture

• Metonymy is a way of establishing a mental address system

• A more salient item (vehicle) is used to access another item (target)

QuickTime™ and aTIFF (Uncompressed) decompressor

are needed to see this picture.

Page 5: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

5

Example 1 of (lexical) metonymy

• We need a good head for this project

(good) headvehicle

part

(smart) persontargetwhole

QuickTime™ and aTIFF (Uncompressed) decompressor

are needed to see this picture.

QuickTime™ and aTIFF (Uncompressed) decompressor

are needed to see this picture.

Page 6: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

6

Example 2 of (lexical) metonymy

• The milk tipped over

milkvehicle

contained

glasstarget

container

QuickTime™ and aTIFF (Uncompressed) decompressor

are needed to see this picture.

Page 7: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

7

Russian example of grammatical metonymy

• брюхан ‘pot-bellied person’

брюхо vehicle

part

брюханtargetwhole

QuickTime™ and aTIFF (Uncompressed) decompressor

are needed to see this picture.

Page 8: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

8

Czech example of grammatical metonymy

• květináč ‘flower-pot’

květinavehicle

contained

květináčtarget

container

QuickTime™ and aTIFF (Uncompressed) decompressor

are needed to see this picture.

QuickTime™ and aTIFF (Uncompressed) decompressor

are needed to see this picture.

Page 9: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

9

Why study grammatical metonymy?

• Grammatical structures are more systematic, more indicative of information structure than lexical structures

• Compare lexical vs. grammatical metonymy

• Compare grammatical metonymy across languages

• May indicate information structure in brain

Page 10: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

10

2. Relevant Previous Scholarship

• Works on metonymy– say almost

nothing about word-formation

• Works on word-formation– say almost

nothing about metonymy

Page 11: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

11

Works on metonymy• Focus on lexical metonymy and on describing

difference between metonymy and metaphor• Jakobson [1956] 1980; Lakoff & Johnson

1980; Lakoff 1987; Langacker 1993, 2009; Croft 1993; Kövecses & Radden 1998; Radden & Kövecses 1999; Seto 1999; Panther & Thornburg 1999, 2002, 2007; Barcelona 2002, Kövecses 2002, Padučeva 2004, Peirsman & Geeraerts 2006

Page 12: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

12

Jakobson [1956] 1980

• Metonymy is based on contiguity.• “Also, as a rule, words derived from the same root,

such as grant -- grantor -- grantee are semantically related by contiguity.”

• “Thus the Russian word mokr-ica signifies ‘wood-louse’, but a Russian aphasic interpreted it as ‘something humid’, especially ‘humid weather’, since the root mokr- means ‘humid’ and the suffix -ica designates a carrier of the given property, as in nelepica ‘something absurd’, svetlica ‘light room’, temnica ‘dungeon’ (literally ‘dark room’).”

• Scholarship has neglected metonymy

QuickTime™ and aTIFF (Uncompressed) decompressor

are needed to see this picture.

Page 13: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

13

Langacker 1993, 2009• “Metonymy is prevalent because our reference-point ability is

fundamental and ubiquitous, and it occurs in the first place because it serves a useful cognitive and communicative function.”

• “By virtue of our reference-point ability, a well-chosen metonymic expression lets us mention one entity that is salient and easily coded, and thereby evoke -- essentially automatically -- a target that is either of lesser interest or harder to name.”

• “Cases where grammatical relationships involve aproximations rather than exact connections, or rely on general or contextual knowledge, are neither atypical nor pathological. ... metonymy in grammar should not be seen as a problem but as part of the solution.”

QuickTime™ and aTIFF (Uncompressed) decompressorare needed to see this picture.

Page 14: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

14

Panther & Thornburg 2002

• Discuss role of metonymy and metaphor in English -er

Padučeva 2004

• Shows that the same metonymic semantic relation can be lexical in one language, but marked by word-formation in another

Page 15: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

15

Peirsman & Geeraerts 2006• Most comprehensive inventory of

metonymy designations

• Focuses primarily on lexical metonymy; grammatical uses do not involve word formation

• Serves as the basis for the system used in my databases

• Will serve as basis for comparisons also (henceforth “P&G”)

Page 16: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

16

Works on word-formation• Mainly lists of suffixes and/or

relationships• 3 Reference Grammars: Švedova 1980,

Dokulil 1986, Faarlund et al. 1997• Šanskij 1968, McFadden 1975,

Maksimov 1975, Rasch 1977, Townsend 1978, Lönngren 1978, Andrews 1996, Janda & Townsend 2000, Townsend & Komar 2000, Araeva 2009

Page 17: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

17

Lönngren 1978• Meanings of suffixes are relations rather than

components, having a converting rather than additive function; 16 are “associative” and 46 are “situative”

Araeva 2009• Mentions metonymy as a possible motive for word formation,

but limited to whole-part/part-whole relationships; her examples are медведь ‘bear’ - медвежатина ‘bearmeat’, горох ‘peas’ - горошина ‘pea’, зверь ‘animal’ - зверье ‘animals’

Page 18: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

18

3. Databases: Russian, Czech, Norwegian

• Based on data culled from Academy/Reference Grammar of each language

• Suffixal word-formation signalling metonymy– includes conversion (zero-suffixation)

• Each database is an inventory of types– no duplicates (examples are merely illustrative!)

Page 19: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

19

A Type is a unique combination of

• Metonymy designation: vehicle & target– брюхан is part-whole– květináč is contained-container

• Word class designation: vehicle & target– both брюхан and květináč are noun-noun

• Suffix: -ан, -áč, etc.

(See sample types on handout)

Page 20: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

20

What the databases do NOT contain

• Word formation that is not metonymical– hypocoristics– caritives– comparative adjectives & adverbs– secondary imperfectives

• Compounding– all types have only ONE root

• Isolated examples, dialectisms• Information on frequency

Page 21: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

21

Challenges in constructing the databases

• Allomorphy or separate suffixes?

• Overlap in metonymies (e.g., part-whole, contained-container, located-location, possessed-possessor)

• Examples with multiple interpretations (e.g., Norwegian maling ‘paint, painting’)

• Extending the P&G inventory to cover all attested types (see next slide)

Page 22: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

22

Vehicles & Targets• Relating to Actions: action, state, change state,

event, manner, time, price-ticketprice-ticket (Czech)• Relating to Participants: agent, product, patient,

instrument• Relating to Entities: entity, abstractionabstraction,

characteristic, group, leaderleader, material, quantity, femalefemale (target only), (target only), malemale (target only) (target only)

• Relating to Part-Whole: part, whole, contained, container, located, location, possessed, possessor

QuickTime™ and aTIFF (Uncompressed) decompressor

are needed to see this picture.

Underlined items have been added More distinctions made within Actions and Participants

Page 23: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

23

The sum is more than the parts

• I do not assume a strict componential analysis via vehicles and targets!

• The unit is the vehicle-target relationship -- a construction that is not just the sum of parts

• Each vehicle-target relationship is unique• For example, action-agent is different from

action-product, not just because of the second member of the relationship

Page 24: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

24

# types

769

576

180

0

100

200

300

400

500

600

700

800

900

Russian Czech Norwegian

Page 25: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

25

# suffixes

284

208

59

0

50

100

150

200

250

300

Russian Czech Norwegian

Page 26: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

26

# metonymy designations

112 109

61

0

20

40

60

80

100

120

Russian Czech Norwegian

Page 27: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

27

Top 13 Metonymy Designations• 10 items found on all 3 top 13 lists:

– abstraction-characteristic– action-abstraction– action-agent– action-characteristic– action-instrument– action-product– characteristic-abstraction– entity-characteristic– characteristic-entity– action-event

action is vehicle for six of them!

Page 28: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

28

Word-class designations

• Vehicles and targets common to all three languages:– adverb, noun, numeral, qualitative

adjective, relational adjective, verb

• Vehicles found only in Russian and Czech:– pronoun, interjection, sound, preposition (R

only).

Page 29: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

29

# word class designations

33

24

12

0

5

10

15

20

25

30

35

Russian Czech Norwegian

Page 30: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

30

Top Ten Word Class Designations

• 8 items found on all 3 top 10 lists:– noun-noun– verb-noun– noun-relational adjective– qualitative adjective-noun– noun-qualitative adjective– noun-verb– verb-qualitative adjective– relational adjective-noun

Page 31: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

31

To what extent does a suffix specify metonymy?

• Number of metonymies per suffix– Highs: 16 (Czech), 15 (Russian), 11

(Norwegian) metonymies per suffix– Lows: only one metonymy for 128 suffixes

(Russian), ... 94 suffixes (Czech), 21 suffixes (Norwegian)

– Average is about 3 metonymies per suffix

• Number of targets per suffix– 60% have only one target, but 15% have

more targets than vehicles

Page 32: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

32

Metonymy designations per suffix

0

20

40

60

80

100

120

140

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

number of metonymy designations

number of suffixes with X metonymy designations

# of R suffixes

# of C suffixes

# of N suffixes

Page 33: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

33

average # metonymy designations per suffix

2.6 2.7

3

0

1

2

3

Russian Czech Norwegian

Page 34: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

34

68%

32%

11%

62.50%

37.50%

12.50%

59%

41%

17%

0%

10%

20%

30%

40%

50%

60%

70%

80%

suffixes with 1metonymy

target

suffixes with>1 metonymy

targets

suffixes withtargets >vehicles

Page 35: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

35

word-class designations per suffix

1.55 1.56 1.61

0

0.5

1

1.5

2

2.5

3

Russian Czech Norwegian

Page 36: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

36

Suffixes and specificity

• Not specific for metonymy

• Target specific for word class

• What does a suffix mean?

• “Given this vehicle X, perform a metonymy such that the target is a member of word class Y.”

Page 37: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

37

4. Observations

• Comparison lexicon vs. word-formation– Metonymy is more diverse and prevalent in word-

formation– But some division of labor between the two

domains

• Directionality– Some metonymies are uni-directional– Most bi-directional metonymies are skewed

• Cross-linguistic comparisons

Page 38: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

38

# metonymy designations

10

36

101

Cited in P&G, notattested in thisstudy

Cited in P&G andattested in thisstudy

Attested only in thisstudy

NOTE: There will be more overlap between P&G andthis study in final version...

Page 39: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

39

Lexicon vs. word-formation

• Some frequent lexical metonymies are not attested in word-formation– agent-product, potential-actual, hypernym-

hyponym

• Some frequent word-formation metonymies are not attested in lexical use– abstraction-characteristic, characteristic-

abstraction, action-abstraction, action-characteristic

Page 40: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

40

59%

41%

62%

38%

62%

38%

0%

10%

20%

30%

40%

50%

60%

70%

bidirectionalmetonymies

unidirectionalmetonymies

Page 41: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

41

Directionality of metonymies in word-formation

• Robust uni-directional metonymies– product-agent, instrument-agent, state-location

• Balanced bi-directional metonymies– entity & characteristic, abstraction & characteristic,

action & product

• Skewed bi-directional metonymies– location-agent, patient-agent, action-agent, action-

characteristic, action-instrument, action-abstraction, action-event, part-whole, contained-container, possessor-possessed, entity-female

Page 42: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

42

Distribution of the 137 metonymy designations by language

52

37

2

2

21

19

5

R, C, NR, CR, NC, NR onlyC onlyN only

Page 43: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

43

Special investments: Russian and Czech

• location-characteristic

• possessor-possessed

• state-characteristic

• characteristic-location

• part-whole

• characteristic-material

Page 44: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

44

Special investments: Russian

• entity-female

• instrument-characteristic

• characteristic-characteristic

Page 45: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

45

Special investments: Czech

• contained-container

• product-location

• quantity-entity

Page 46: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

46

Special investments: Norwegian

• location-located

• product-agent

Page 47: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

47

5. Conclusions

• The main purpose of word-formation is to signal metonymy

• Metonymy in word-formation is more diverse than in lexical use

• Different languages make different investments in word-formation to signal metonymy

• Compare lexical vs. grammatical systems of meaning (Talmy 2005)

Page 48: “Building Words via Metonymy: A Comparison of Russian, Czech and Norwegian” Laura A. Janda Universitetet i Tromsø

48

Metonymy continuumLE

XIC

AL

conv

ersi

on

HIG

H m

eton

ymy/

suff

ix

1 m

eton

ymy/

suff

ix

com

poun

ding

......................# met/suffix