Cogl.2005.16.1 Ozacaliskan Motion Events Metaphor

8/3/2019 Cogl.2005.16.1 Ozacaliskan Motion Events Metaphor

http://slidepdf.com/reader/full/cogl2005161-ozacaliskan-motion-events-metaphor 1/40

Metaphor meets typology: Ways of movingmetaphorically in English and Turkish

SEYDA OZCALISKAN*

Abstract

Earlier work on literal motion has shown that English and Turkish belong

to typologically distinct classes of languages, with English speakers paying

greater linguistic attention to the manner dimension of motion events (e.g.,

O zcalıskan and Slobin 1999a, 2003). As a further step, this article inves-

tigates whether typological di¤erences hold true for the metaphorical

extensions of motion events. Thus, the article compares two types of lan-

guages with regard to their lexicalization patterns in encoding metaphorical

motion events: (1) verb-framed languages (or V-languages, represented by

Turkish), in which the preferred pattern for framing motion events is the

use of a path verb with an optional manner adjunct (e.g., enter running),

and (2) satellite-framed languages (S-languages, represented by English),

in which path is lexicalized in an element associated with the verb, leaving

the verb free to encode manner (e.g., run in). The analysis of written texts

and elicited responses in the two languages shows clear typological contrast,

with English speakers encoding manner of motion in their metaphorical de-

scriptions more frequently and extensively than Turkish speakers, using a

variety of linguistic devices (e.g., verbs, adverbials). Overall, the resultsindicate that the degree of codability of a semantic dimension in a lexical

item (i.e., motion verb) has a spillover e¤ect on the choice of other lexical

items in a sentence, suggesting the conceptual salience of this dimension for

its speakers. This e¤ect is observable in both the literal and the metaphori-

cal uses of the lexicon.

Keywords: motion events; metaphorical motion; verb-framed typology;

satellite-framed typology; manner of motion.

Cognitive Linguistics 16–1 (2005), 207–246 0936–5907/05/0016–0207

6 Walter de Gruyter



1. Introduction

Earlier theoretical and experimental work on conceptual metaphor theory

has shown that both English and Turkish pervasively structure a widerange of abstract domains using motion in space, including time (e.g., mo-

ments slip by), ideas (e.g., the idea sprang back into his mind ), emotions

(e.g., she felt a sudden surge of emotions), the economy (e.g., prices plum-

met, inflation skyrockets), and so forth (Lako¤ and Johnson 1980, 1999;

Ozcalıskan 2002, 2003a, 2003b). Furthermore, theoretical work on the

linguistic organization of motion events has shown it to be a domain

that can be construed in radically di¤erent ways in di¤erent languages,

but which at the same time can be described by a limited set of under-

lying universal patterns (Talmy 1985, 2000). As proposed by Talmy, theworld’s languages can be grouped into a two-category typology in terms

of the way the core feature of a literal motion event—which is the path of

motion —is expressed linguistically, with some languages encoding this

feature in the verb, and others in a satellite to the verb (a particle or a pre-

fix). Talmy refers to these two types as verb-framed and satellite-framed

languages (or V-languages and S-languages), respectively.1 This article fo-

cuses specifically on the metaphorical extensions of motion events (e.g.,

hours crawl by, the idea bounces back into my mind ), and compares the

lexicalization patterns of two typologically distinct languages, English(an S-language) and Turkish (a V-language), in encoding various seman-

tic components of a metaphorical motion event. The article investigates

whether the typological di¤erences extend to the metaphorical uses of

spatial motion, and if so, what consequences this has for our conceptual-

ization of various target domains that are structured by the source do-

main of motion in space.

1.1. Outline of the typology

As proposed by Talmy, path of motion constitutes the core feature of a

motion event, and languages show two distinct lexicalization patterns by

typically encoding path of motion in either a verb (e.g., exit, ascend ) or

an associated satellite (e.g., go out, go down). Motion events also di¤er

in terms of their degree of structural complexity. Thus, one can di¤erenti-

ate between a unitary event (e.g., he went into the room) and a complex

one (e.g., he crawled into the room). A unitary event indicates only one

dimension of motion, which, in this case is the path information (into).

On the other hand, a complex event encodes both the manner (crawling )and the path (into) components of a motion event within a single clause

(Talmy 2000). Languages di¤er in their ways of expressing the compo-

nents of a complex event, with S-languages typically conflating manner

208 S. O zcalıskan



with motion, and V-languages conflating path with motion in the main

verb of a clause. The di¤erence in conflation patterns has significant

e¤ects on the relative codability of the semantic domains that con-

stitute the components of a motion event. Since S-languages prefer toencode path using satellites, the main-verb slot becomes available for a

manner verb (e.g., walk /run/crawl ... in/out/across ...).2 This provides

S-language speakers with a more accessible and easily codable linguistic

option for indicating manner of motion. As a consequence, S-language

speakers encode manner habitually, develop a richer lexicon of manner

verbs, and make finer lexical distinctions within the domain of manner

(Slobin 2000, 2003). By contrast, in V-languages, the main verb is chiefly

reserved for encoding path information, and there is no other easily cod-

able linguistic slot with which to encode manner of motion. Therefore, in

contexts where attention to manner is salient, V-language speakers typi-

cally rely on subordinated manner verb constructions (e.g., enter/exit by

running ) to indicate manner, but due to the relative syntactic complexity

of subordinated expressions, manner information is omitted in most in-

stances in V-languages.3

The typological contrast becomes particularly salient with events that

involve the crossing of a boundary (e.g., motion into or out of a bounded

space), where the lexical constraints on encoding manner in the main verb

are stricter for V-languages. Thus—as depicted in Figure 1—the pre-

ferred patterns for describing a simple motion event which involves both

manner (running) and path (interior of the house as the goal of move-

ment) components will be such that English speakers will choose to

encode both manner and path by conflating motion with manner in the

main verb and indicating path with the particle into as in he ran into the

house. Turkish speakers, on the other hand, will typically encode only

path by conflating motion with path in the main verb and leaving out

manner information, as in eve girdi ‘he entered the house’. However, ininstances where manner becomes perceptually salient, Turkish speakers

may choose to encode manner as well, typically by subordinating manner

Figure 1. A typical motion event

Moving metaphorically in English and Turkish 209



to the main path verb of a clause (eve kosarak girdi ‘he entered the house

running’). These linguistic di¤erences, in turn, are likely to have e¤ects on

the organization of mental representations, leading to di¤erent mental

imagery regarding how one navigates in space (Slobin 1997, 2000, 2003).Speakers of English have linguistic access to a richer array of motion

events that involve manner due to the high codability of this dimension

in their native language. Therefore, compared to Turkish speakers, En-

glish speakers may be more likely to pay greater linguistic attention to

and detect more fine-grained variations in the manner dimension of

motion events, which in turn may increase the conceptual salience of this

dimension for them.4

Empirical research based on analysis of written texts and orally eli-

cited narratives from child and adult native speakers has provided con-

vincing evidence for the proposed typological di¤erences, with clear indi-

cations of di¤erential linguistic attention paid to manner of motion by

speakers of these two groups of languages for literal motion events (e.g.,

Ibarretxe-Antunano 2001; Naigles et al. 1998; Oh 2003; Ohara 1999; Pa-

pafragou et al. 2002; Ozcalıskan and Slobin 1999a, 1999b, 2000a, 2000b,

2003; Slobin 1996, 1997, 2000). The studies have shown higher frequency

of mention and greater lexical diversity with regard to the manner com-

ponent of motion events by S-language speakers. The typological contrast

seems to have even wider applicability across various other languages of

the world, and to crosscut boundaries of culture, language family and

geographical location (Slobin 1997, 2003).

This article investigates the applicability of the proposed distinctions

of the typology to metaphorical extensions of motion events, in a com-

parison between English (S-language) and Turkish (V-language). Meta-

phor is defined as a conceptual-linguistic mapping between two conceptual

domains: the source domain, which serves as the source of vocabulary and

conceptual inferences, and the target domain, to which vocabulary and in-ferences are extended metaphorically (Lako¤ and Johnson 1980, 1999).

The article focuses on metaphors that use motion in space as the source

domain. Metaphorical motion is defined as any verb of motion—self or

caused—with a nonmotion interpretation and metaphorical change of lo-

cation.5 The article focuses only on metaphorical motion events, and all

the tables and figures presented in the results section refer only to such

metaphorical uses. Nonmetaphorical (i.e., literal) motion events (e.g., he

ran into the house) were not included in the analysis.

The study involves both descriptive (analysis of di¤erent types of written texts) and experimental (structured elicitations from adult na-

tive speakers) data collection techniques. It aims to reveal the di¤erences

between the two languages in encoding the manner component of meta-

210 S. O zcalıskan



phorical motion events. The expectation is twofold: On the one hand—in

line with the results of earlier work on literal motion—metaphorical mo-

tion-event descriptions in S-languages (here, English) are expected to in-

volve finer lexical distinctions at the level of manner of motion, with agreater amount and diversity of manner verb use. On the other hand,

one might argue that V-language speakers (here, Turkish) may have re-

course to other means of encoding manner of motion in addition to man-

ner verbs. Lexical means such as adverbials of manner (hesitantly, rap-

idly), descriptions of the physical or inner state of the moving entity ( he

was exhausted ) or of the terrain of movement (the road was slippery) can

also serve as tools for conveying manner information. Given the equal ac-

cessibility of such alternative lexical means of encoding manner for both

language types, Turkish speakers may use these means to encode manner

that is not easily expressed at the level of motion verbs, and thereby con-

vey manner information at comparable rates to English speakers.

Statements of metaphorical mappings are presented in small capitals

throughout the text, and the linguistic metaphors in the examples

are underlined, unless indicated otherwise. For the examples in Turkish,

morpheme-by-morpheme glosses are only provided for the underlined

segments of the excerpt, together with a free translation of the full ex-

cerpt. An explanation of the abbreviated labels for the morphemes is pro-

vided in the Appendix.

2. Sample

2.1. Written texts

The sample included twenty novels, ten in each language, and five widely

read daily newspapers in each language. The novels included works of

both contemporary and earlier writers, and an e¤ort was made to includenovels that were richer in metaphorical motion events in both languages.

2.2. Elicitations

The sample included twenty adult native speakers of English and twenty

adult native speakers of Turkish. The English data were collected in

Berkeley, California and the Turkish data were gathered in Istanbul. The

participants ranged in age from 23 to 63, with a mean age of 31 for theTurkish sample and 33 for the English sample. An attempt was made to

include people with di¤erent backgrounds including editors, college pro-

fessors, writers, architects, and graduate students in various social science




fields, in order to take into account the variation in the metaphorical uses

of motion events.

3. Procedure

3.1. Procedure for data collection

3.1.1. Written texts Each novel was opened randomly ten times, and

at each opening the first five instances of metaphorical motion events

were recorded, resulting in about 50 such instances from each novel.

Each newspaper was followed for three consecutive days. On each day,

all the opinion sections (i.e., editorials, columns) and a front-page news

story were extracted, and all instances of metaphorical motion eventswere recorded.6

3.1.2. Elicitations Each consultant was interviewed individually and

given a two-page questionnaire in his/her native language that asked for

metaphorical motion descriptions for various target domains. The first

page of the questionnaire included instructions, along with an example

response. The instructions were as follows:

We human beings have the tendency to formulate abstract concepts in concrete(or physical) terms. We think of arguments as wars (e.g., he battled over his

ideas), emotional states as locations (e.g., he fell in love), theories as buildings

(e.g., he has a solid argument), and so forth. Similarly, our everyday thinking has

the tendency to conceptualize an abstract set of things (e.g., love, life, conscious-

ness) in terms of a concrete or physical set of things or events. For instance, we

think of human personality types in terms of temperature, and talk about a person

as being a warm person, a cold person, having a chilly personality, and so forth.

Similarly, we conceptualize our romantic a¤airs in terms of a journey, and talk

about our relationship as not going anywhere, hitting a dead-end street, running

pretty smoothly, and so forth. In the next page, you will find a set of questions

that involve such abstract concepts. The task at hand is to think of all these

abstract concepts in terms of the motion of a moving entity—this entity can be a

human being, an animal, a river, wind, or any moving artifact such as a car, a

plane, a boat—and come up with descriptions for each one of them. Think of

this as a creative exercise, but limit yourself to the possibilities available in your

language.

The second page of the questionnaire included the entries to be filled in.

There were two forms, A and B, with five di¤erent target conceptual do-mains in each form. Each subject received one of the two forms. The

choice of target domains was based on the analysis of written texts, which

revealed that the domains of mental/emotional states, time, death, life,

212 S. O zcalıskan



and sickness were described most extensively in terms of motion in space

in both languages. For each domain, the consultant responded to the fol-

lowing question:

If you think of DOMAIN X as a moving entity, what are the di¤erent movements

that it could make? List the first five motion verbs that come to your mind.7

Subjects faced no time limits in filling out the questionnaire.

3.2. Procedure for data analysis

Metaphorical motion events (e.g., the idea races through her mind like a

flame, he drifts into a state of utmost solitude, the ambiguity in her voice pulls his worst fears to the surface) were taken as the unit of analysis, and

the two languages were compared in terms of the extent to which they

encoded manner information. The analysis involved quantitative and

qualitative comparisons of the English and the Turkish data in terms of

various linguistic categories that encoded manner of motion, including

verbs and nominalized verbs of motion, aspectual marking on the verbs,

and adverbials, adjectives, and verb complements qualifying verbs of

motion. Motion verbs were divided into three categories (manner, path,

neutral), based on the information encoded in the verb. A list of motionverb types included in the analysis is provided below.8

i. Manner verbs (V-manner), e.g., run, fly, jump, plunge, drag, launch,

push, creep.

ii. Path verbs (V-path), e.g., enter, exit, ascend, descend, follow,

approach, withdraw.

iii. Verbs with no manner or path (V-neutral), e.g., go, move.

4. Findings9

4.1. Written texts

4.1.1. Verbs of motion Analysis of verbs and verbal nouns of meta-

phorical motion suggests a clear typological dichotomy. As can be seen in

Figure 2, novels written in English include a significantly (w2ð1Þ ¼ 18:05,

p < :001) higher percentage of manner verbs (59%) than novels written

in Turkish (21%). Turkish novels, on the other hand, mainly rely on bare

path verbs (71%) in describing metaphorical motion events.A few examples from the data are provided to illustrate the di¤erence

between the two languages. All the examples show a clear preference for

manner verbs in English (clamber, creep, whirl, fly), and path verbs in




Turkish ( pass, come, enter, approach). The preference is clearly markedacross conceptual domains that constitute the target domain of the meta-

phorical mappings. Each of the following metaphor pairs map onto dif-

ferent target domains, including the body ([1] and [2]), death and sickness

([3] and [4]), and emotional states ([5] and [6]). However, regardless of

the target domain, English writers consistently prefer manner verbs, while

Turkish speakers choose non-manner verbs.

(1) She had the impression now that he had clambered back inside him-

self and shut the door. (Bowles 1966: 182)(2) ‘‘I stanbul’u

Istanbul-ACC

dinliyorum,’’

listen-PRESENT-1SG

diye gecirdi

pass-CAUS-PAST

icinden.

interior-POSS:3SG-ABL

‘‘Beyog ˘ lu artık I ˙stanbul deg ˘ il. I ˙stanbul

arka sokaklara kacıstı.’’ (Mungan 1993: 133)

‘‘‘I am listening to Istanbul’’, he let it pass from his inside.

‘‘Beyoglu is not Istanbul anymore. Istanbul ran to the back

streets.’’ ’

(3) The town was a shambles; corpses, mangled by butchers andstripped by plunderers, lay thick in the streets; wolves sneaked from

the suburbs to eat them; the black death and other plagues crept in

to keep them company, and the English came marching on; while

7%

59%

34%

8%

71%

21%

0

10

20

30

40

50

60

70

80

90

100

manner path neutralverb type

p e r c e n t v e r b u s e

EnglishTurkish

Figure 2. Percentage of metaphorical motion verbs in novels (computed by dividing the total

number of manner, path, or neutral verbs by the total number of motion verbs in

each language)

214 S. O zcalıskan



the danse macabre whirled about the tombs in all the cemeteries . . . .

(Miller 1961: 43)

(4) Bu ilk karsılasmamız deg ˘ ildi. O lu mle

death-INS

ilk

first

yu z

face

yu ze

face-DAT geldig ˘ imde

come-PAST-NOM-1SG

do rt yasındaydım. Bir dag ˘ mantarının

cekici kılıg ˘ ına bu ru nmu s olarak sokuldu

approach-PAST

yanıma. (Aral 1999: 23)

side-POSS:1SG-DAT

‘This was not our first encounter. When I first came face to face

with death I was four years old. It approached me disguised in the

attractive outfit of a mountain mushroom.’

(5) When Evelyn learned of my pregnancy with little Pierre, as Adam

and I and my parents used to call him, she flew into a rage that sub-

sided into a years-long deterioration and rancorous depression.

(Walker 1993: 127)

(6) Yu zlerde okuyacag ˘ ı ve yavas

slow

yavas

slow

icine

interior-POSS:3SG-DAT

gireceg ˘ i

enter-FUTURE-NOM-POSS:3SG

dehsete

terror-DAT

iste bo yle bo yle

yaklasmaya

approach-NOM-DAT

basladı. (Pamuk 1996: 284)

start-PAST

‘He started approaching the terror just like this, the terror that he

would read in the faces of others and he himself would slowly enter

into.’

The patterns are found to be the same in the newspapers. As in the case

of novels, newspapers in English include a significantly (w2ð1Þ ¼ 13:9,

p < :001) higher percentage of manner verbs (50%) than newspapers in

Turkish (19%). Turkish journalists, on the other hand, mainly rely onbare path verbs in their metaphorical motion-event descriptions (76%;

see Figure 3).

A few excerpts from the newspaper articles in the two languages are

presented below. Similar to novelists, journalists writing in English rely

mainly on manner verbs in their metaphorical descriptions that tap into

a wide variety of target domains, including economy, politics, foreign af-

fairs, demographics, and so forth.

(7) The human race has never had it so good. Per capita wealth hadsoared in this century. Frightful diseases have been eradicated.

Infant mortality has plummeted. (Je¤ Jacoby, Boston Globe, 14

October 1999)




(8) . . . Bradley is surfing a wave that threatens to capsize Vice Presi-

dent Al Gore. (Commentary, LA Times, 30 September 1999)

(9) Whereas Bluefields children somersault through complex sentences,

Winston lumbers along slowly, one sign at a time. (Lawrence Os-

borne, NY Times Magazine, 25 October 1999)

(10) The economy continues to steam ahead and inflation continues to

lie on the floor . . . ( Robert Dodge, Dallas Morning News, 29 Octo-

ber 1999)

The pattern is reversed for the Turkish newspapers, in which case journal-

ists mostly use bare path verbs to talk about similar issues:

(11) Ekonomiye

economy-DAT

destek

support

paketinden

package-POSS:3SG-ABL

beklenen

expected

vergi

tax

yumusaması

soften-NOM-POSS:3SG

cıkarsa

exit-COND

ve buna bag ˘ lı

olarak piyasaya

market-DAT

para

money

girisi

enter-NOM-POSS:3SG

devam ederse yeni bir tarihi zirveye dog ˘ ru hareket yaratılabilir.

(Esin Cetinel, Posta, 19 July 1999)

‘If the expected tax cuts exit from the economy support package,and in relation to this, if the entry of money to the money market

continues, a movement can be created towards a new historic

summit.’

9%

50%

41%

5%

76%

19%

0

10

20

30

40

50

60

70

80

90

100

manner path neutralverb type

p e r c e n t v e r b u s e

EnglishTurkish

Figure 3. Percentage of metaphorical motion verbs in newspapers (computed by dividing the

total number of manner, path, or neutral verbs by the total number of motion verbs

in each language)

216 S. O zcalıskan



(12) Bunlar ‘‘su re’’ yiyecekler. Tuhaf bir sekilde hayatına

life-POSS:2SG-DAT

girip

enter-CONV

bir daha do nmemek

return-NEG-INF

u zere

so.as

cıkıp

exit-CONV gidiyorlar.

go-PRESENT-PLU

Bazı insanlar gibi. Bir de tabii hic

never

hayatından

life-POSS:2SG-ABL

cıkmayacak

exit-NEG-FUTURE

olanlar

be-NOM-PLU

var. Patates gibi, kalkan balıg ˘ ı gibi . . . (Perihan Magden, Radikal ,

exist

21 July 1999)

‘These are ‘‘temporary’’ food. They enter one’s life in a strange

way, and exit and go without any promise of return. Like somepeople. And, of course, there are the ones that will never exit from

one’s life. Like potatoes or shield fish . . .’

I should add a word of caution here about encoding of path in English.

Even though English writers do not typically encode path of motion

in the verb, they nevertheless include detailed path information in their

metaphorical descriptions using directional particles and prepositions

(Ozcalıskan 2004). In fact, most verbs of motion in English are accom-

panied by at least one directional particle that conveys path informa-tion (see examples [13]-to-[15]; directional particles and prepositions are

underlined).

(13) Her mind stumbled backwards to other embraces of his, and back

all the way to that night in Florida years ago. (Oates 1967: 244)

(14) Then she fell back almost voluptuously into a world of undi¤eren-

tiated flapping things where words were silent and colors became

textures. There were blossomings and explosions. From where she

had floated far down the coastline of her consciousness, she calledout. (Bowles 1966: 116)

(15) It is about the demise of the era of big-city bosses, the way televi-

sion changed electoral politics, how the Irish in America climbed

out of their immigrant poverty on a ladder of politics . . . (Je¤

Jacoby, Boston Globe, 25 October 1999)

The typological contrast is also quite evident in the diversity of the

manner verb lexicon. As can be seen in Table 1, novels written in English

contain three times as varied a manner lexicon as Turkish novels (95 to 30types). Similarly, newspapers in English contain more than three times as

varied a manner lexicon as their Turkish counterparts (110 to 32 types).

The di¤erence becomes even more pronounced when the verbs from




novels and newspapers are combined together, with texts in English hav-

ing four times as varied a manner lexicon as the Turkish texts (145 to 41

types; w2ð1Þ ¼ 58:15, p < :001). Lists of all the manner verbs observed

in novels and newspapers in the two languages are provided in the

Appendix.

Typically, for a single verb in Turkish that describes a motion with

manner, English has at least two or more verbs that describe the same

motion event. An illustration of this phenomenon is provided in Table 2

with examples from the data.

It is evident from this summary that English makes finer distinctions

within domains of metaphorical motion that involve manner. This ex-

tensive di¤erentiation becomes especially striking in the variety of verbs

that are used in English to convey particular motor patterns such as walk-

ing and running. For the single Turkish verb, yu ru ‘walking’, English

texts use twenty-three di¤erent verbs (e.g., walk, drift, lumber, meander,

stride, trot). Similarly, for the single Turkish verb, kos ‘running’, English

texts employ nine di¤erent verbs (e.g., run, flee, flit, race, charge), all of

which encode nuances on a basic motor pattern of running. In summary,

the analysis strongly supports the typological dichotomy between the two

languages in encoding manner of motion. Texts written in English includea greater frequency and diversity of manner verbs than texts written in

Turkish, and this di¤erence is marked in both novels and newspapers.

Interestingly, however, the analysis of verbs and nominalized verbs of

metaphorical motion in the two languages has also revealed the e¤ect of

an intervening variable, and this variable is narrative perspective. In both

languages, the shift in narrative perspective has an e¤ect on verb choice,

leading to a di¤erence between self-motion and caused-motion descrip-

tions. Novels and newspapers produced in both languages include a

higher percentage of manner verbs when the metaphorical motion eventsare rendered from a caused-motion perspective as opposed to a self-

motion perspective. As can be seen in Table 3, novels in English use al-

most twice as many manner verbs (94% to 55%), and novels in Turkish

Table 1. Frequency distribution of types of manner verbs in English and Turkish

English Turkish

Novels 95 30 (36)aNewspapers 110 32 (40)

Novels and newspapers combined 145 41 (53)

a Numbers in parentheses refer to counts that also include verbs with the same roots, but

with derivational su‰xes (e.g., atla-t ‘jump-CAUSATIVE su‰x’) counted as additional

verb types.

218 S. O zcalıskan



include three times as many manner verbs in the caused-motion descrip-

tions as compared to the self-motion descriptions (45% to 15%). These

di¤erences are found to be statistically significant at the w2ð1Þ ¼ 10:20,

p < :01 level for English, and w2ð1Þ ¼ 15:0, p < :001 for Turkish.

Similarly, as can be seen in Table 4, newspapers in English contain al-

most twice as many (79% to 42%) manner verbs when the event perspec-

tive is one of caused motion as opposed to self-motion. The pattern is the

Table 2. Examples of Turkish and English verbs of motion encoding manner

Turkish English

ak ‘flow’ flow, streamdo k (u l ) ‘pour’ pour, spill, slop

sız ‘leak’ leak, drain

kay ‘slide’ slide, slip

yuvarlan ‘roll’ roll, tumble, wallow

su ru n ‘crawl’ creep, crawl

tırman ‘climb up’ climb, clamber, skyrocket, soar

sendele, to kezle ‘stumble’ falter, stagger, stumble, trip

atla, sıcra ‘jump, bounce’ bounce, bound, jump, plummet, skip, spring, scramble

atıl ‘leap’ leap, lunge, lurk, launch, swoop

fırla ‘dart’ dart, burst, bolt, surge, pop, spurt

dal ‘plunge’ dip, plunge

su r ‘ride, drive’ ride, drive

cek ‘pull’ draw, pull

it ‘push’ push, propel, shove

yu z ‘swim’ float, flood, swim

kos ‘run’ run, flee, fleet, rally, race, reel, surge, charge, flit

kovala ‘chase’ chase, pursue, track, trail

yu ru ‘walk’ walk, drift, ebb, flounce, linger, lumber, march, meander, roam,

rustle, stride, tread, worm one’s way, hike, pace, ramble,

snake, trample, trot, swarm, forge, hurry, rush

Table 3. Percentage of metaphorical motion verbs in novels by narrative perspectivea

English Turkish

self caused self caused

Manner verbs 55% 94% 15% 45%

Path verbs 38% 5% 75% 55%

Neutral verbs 7% 1% 10% 0%

Total 100% 100% 100% 100%

a Percentages were computed by dividing the total number of manner, path, or neutral verbs

by the total number of motion verbs separately for each language and narrative perspective.




same in the Turkish newspapers, which include four times as many man-

ner verbs in the caused-motion descriptions as opposed to the self-motion

descriptions (45% to 11%). These di¤erences are found to be significant

for both English (w2ð1Þ ¼ 11:31, p < :001) and Turkish (w2ð1Þ ¼ 20:64,

p < :001).

A few examples are provided to give the reader a sense of the di¤er-

ences between self-motion and caused-motion descriptions in the two lan-

guages. The first in each pair describes a metaphorical motion event from

a self-motion perspective, and the second describes a similar motion eventfrom a caused-motion perspective. As examples (16) to (18) illustrate, in

both languages self-motion descriptions (16a, 17a, 18a) are less likely to

elicit manner verbs than caused-motion descriptions (16b, 17b, 18b).

(16) a. He could hear the words coming out of his mouth, but even

as he spoke them, he felt they were expressing someone else’s

thoughts . . . (Auster 1990: 36)

b. She made me tell her the whole story of my relations with

Georgie in detail; and . . . I poured it all out with relief, and asI talked Antonia held my hand. (Murdoch 1976: 99)

(17) a. ‘‘. . . sizi memnun edecek birtakım sonuclara

conclusion-PLU-DAT

varacag ˘ ınızı

arrive-FUTURE-2PL-NOM

umuyorum.’’ (Cuneyt

Arcayurek, Cumhuriyet, 27 July 1999)

‘ ‘‘. . . I am hoping that you will arrive at some conclusions

that will satisfy you.’’’

b. Tarihi historical

kisiliklerpersonality-PLU

bazensometimes

du su ndu klerinden

cok

very

farklı

di¤erent

sonuclara

consequence-PLU-DAT

Table 4. Percentage of metaphorical motion verbs in newspapers by narrative perspectivea

English Turkish

self caused self caused

Manner verb 42% 79% 11% 45%

Path verb 47% 18% 82% 55%

Neutral verb 11% 3% 7% 0%

Total 100% 100% 100% 100%

a Percentages were computed by dividing the total number of manner, path, or neutral verbs

by the total number of motion verbs, separately for each narrative perspective and language.

220 S. O zcalıskan



dog ˘ ru

toward

su ru klenir, hic akıllarına gelmeyen

drag-CAUS-PRESENT

gelismelere yol

acarlar. (Turker Alkan, Radikal , 20 July 1999)

‘Historical personalities are sometimes dragged towardsconsequences very di¤erent from what they have ever thought

about, they open the way for developments that never come

to their minds.’

(18) a. September’s decrease in durable goods orders was led by a 3.9

percent drop in orders for transportation equipment . . . . ( News

report, New York Times, 27 October 1999)

b. Slackening demand for cars and airplanes in September helped

drive down overall orders for big ticket manufactured goods by

1.3 percent. (News report, New York Times, 27 October 1999)

In summary, the preference for describing a metaphorical motion

event from a caused-motion perspective leads to greater manner verb use

in both English and Turkish, a di¤erence also observed in literal-motion

events (Ozcalıskan 2000). One possible explanation for this is that caused-

motion perspective necessitates a stronger evaluative component than a

self-motion perspective, because it gives cues about narrative motivation

in a particular act and its associated outcome, leading to greater use of

manner verbs. However, even though Turkish writers employ more man-ner verbs when they utilize a caused-motion perspective, the typological

di¤erence in terms of the frequency of manner verb use is still marked be-

tween the two language types. That is, even when considering transitive

and intransitive verbs separately, English still uses manner verbs at a

higher rate than Turkish (in novels: self-motion 55% versus 15%, caused

motion 94% versus 45%; in newspapers: self-motion 42% versus 11%,

caused motion 79% versus 45%). Therefore, it is safe to claim that lex-

icalization preferences associated with the choice of a particular narrative

stance influence the extent to which the typological dichotomy exerts it-self, but do not override it.

4.1.2. Verbs of motion: subordination of manner As pointed out earlier,

encoding manner information in a V-language typically involves heavier

syntactic packaging (e.g., subordinate constructions). In most instances,

the syntactic overload renders this option less likely to be used by V-

language speakers. In fact, no instances of subordinated manner-verb

constructions are observed in the metaphorical uses of motion events in

the Turkish data, neither in the novels nor in the newspapers. Subordi-nated manner expressions are also found to be very rare in the English

data. There are only two instances of such, both of which come up in the

novels:




(19) The days passed, and even though there was rarely a moment when

they were not together, he continued to say nothing about what

truly concerned him—nothing about the struggle to put his life

together again, nothing about how he saw the wall as a chance toredeem himself in his own eyes, nothing about how he welcomed

the hardships of the meadow as a way to atone for his recklessness

and self-pity—for once he got started, he knew that all the words

would come tumbling from his mouth, and he didn’t want to

make Pozzi any more nervous than he already was. (Auster 1990:

127)

(20) I let her take me to places I’ve never seen before, the swell bars

around the Champs-Elysees where the sound of jazz and baby

voices crooning seems to soak right through the mahogany wood-work. Even when you go to the lavabo these pulpy, sappy strains

pursue you, come floating into the cabinet through the ventilators

and make life all soap and iridescent bubbles. (Miller 1961: 177)

This finding presents an interesting contrast to Spanish, which is also a

V-language. Analysis of written and oral narratives in Spanish has shown

that Spanish speakers make heavier use of subordinated manner expres-

sions than Turkish speakers (Ozcalıskan and Slobin 1999a, 1999b). This

intra-typological di¤erence can be attributable to the word-order e¤ect.In a typical clausal construction in Turkish, the main verb comes at the

end of the sentence, and any verb subordinated to the main verb must

precede this main verb. Therefore, Turkish speakers have to make a deci-

sion to include a subordinated manner verb well in advance, before they

produce the main verb (e.g., Eve kosarak girdi ‘house-to by-running

entered-he’; ‘He entered the house at a run’). On the other hand, Spanish

speakers can add subordinated manner verbs ad hoc, after they have pro-

duced the clause in its full form (e.g., El e ntro en la casa corriendo ‘He

entered the house running’), resulting in a greater use of such subordi-

nated manner verb constructions in Spanish.

4.1.3. Verbs of motion: Aspectual marking and repeated verbs The in-

flectional morphology of Turkish provides one relatively easy, but some-

what constrained option of marking manner on the verb by means of as-

pectual su‰xes attached to the motion verb root. A purely directional

verb such as gir ‘enter’ with an aspectual su‰x -iver ( gir-iver ‘enter-give’)

conveys manner information (suddenness, abruptness) along with path.However, this aspectual su‰x is very limited in its use; it only encodes

information about the rate of motion by adding a sense of ‘‘suddenness’’

to whatever verb it is attached to. A search of the data for this aspectual

222 S. O zcalıskan



marking has revealed only two instances, pointing to its limited use by

native speakers of Turkish.

(21) Umberto Eco’nun vurgulaması, bu haftaki Time dergisininkapag ˘ ıyla u stu ste gelince

bizim

our

ko semizde

column-POSS:1PL-LOC

de sag ˘ lık

health

konusu

subject-POSS:3SG

o ne

front-DAT

cıkıverdi. (Zulfu Livaneli, Sabah, 14 July 1999)

exit-give-PAST

‘When Umberto Eco’s emphasis overlapped with the cover story

of this week’s Time magazine, the subject of health suddenly

exited to the front in our column.’

(22) . . . Ecevit, alınan son ekonomik kararlarla sermaye

investment

du zeninin

sector-POSS:3SG-POSS:3SG

umudu

hope-POSS:3SG

haline

state-POSS:3SG-DAT

geliverdi. (Cumhuriyet, July 28, 1999,

come-give-PAST

Sadullah Usumi)

‘. . . with the latest economic decisions, Ecevit suddenly became the

hope of the investment sector.’

English, on the other hand, due to the lack of a rich inflectional mor-

phology, relies on di¤erent linguistic means to create somewhat similar

e¤ects. Especially novels written in English make use of repeated motion

verbs (e.g., run and run) and aspectual verbs (e.g., keep slipping ) in at-

tempts to amplify the manner component of a metaphorical motion

event, as in examples (23) to (25).

(23) She could feel him ebbing away, ebbing away, leaving her there like

a stone on a shore. He was withdrawing [mentally], his spirit wasleaving her. (Lawrence 1980: 185)

(24) He struggled to settle down again, but his mind kept wandering

back to the road, to the exhilaration he had felt for those two

weeks, and little by little he began to give himself up for lost.

(Auster 1990: 8)

(25) She felt her heart sinking just at the touch of him. Her mind kept

slipping o¤ the wedding ceremony and onto Lowry, imagining

him getting out of his car and going to the doorway, climbing the

stairs, and not finding her home. (Oates 1967: 167)

Turkish, on the other hand, does not make any use of repeated or as-

pectual verbs. There is only one instance of a repeated noun use, found in




a novel, which indirectly adds manner to the path verb gel ‘come’. In this

example, the repetition of the word dalga ‘wave’ clearly conveys an in-

crease in force at every successive arrival of the wave:

(26) I cinden

inside-POSS:3SG-ABL

o nce

first

bir

one

sevinc

joy

dalgası

wave-POSS:3SG

kabarır gelir,

come-PRESENT

bu belli belirsizdir, sonra

then

bir

one

dalga,

wave

bir

one

dalga

wave

daha . . . (Kemal 1997: 248)

more

‘A wave of joy foams up and comes from his inside, at first this is

almost unnoticeable, then another wave, another wave . . .’

In summary, as the foregoing analysis suggests, the semantic domain of

manner of motion is much less di¤erentiated in Turkish. The di¤erence is

marked in terms of both the frequency of manner verb use and the rich-

ness of the manner verb lexicon. Apart from manner verbs, there are a

few other options Turkish speakers can use to mark manner on the verb,

either via subordinated manner verb constructions or aspectual su‰xes

attached to the verb root. However, as observed in the data, both of these

options are either completely disregarded or used very rarely by Turkish

native speakers. Manner is likely to constitute a perceptually salient di-

mension of a motion event for speakers of any language. However, as

the above analysis has shown, the encoding of manner information relies

heavily on codability (Slobin 2004), and speakers of a language are

more likely to pay attention to and elaborate manner if they use a lan-

guage in which this dimension is highly codable. As discussed earlier,

compared to Turkish, English allows for easier codability, where manner

can be expressed using a single, finite, high-frequency lexical item (i.e., a

main verb) rather than a phrase, or a nonfinite verb such as a subordinate

construction (Slobin 2004). This allows English speakers to encode man-ner habitually, develop a richer lexicon of manner verbs, and make finer

distinctions at the predicate level within the domain of manner.

This brings us to our second question. If Turkish does not provide a

slot for marking of manner information that is easily codable, and if the

only available option is to use a construction type that is syntactically

complex and speakers prefer not to use this option, then what do Turkish

speakers do? Do they leave out manner information in their description

of motion events, or do they use alternative lexical means of encoding

this information? I will address this issue in the next section.

4.1.4. Alternative lexical means There are, in fact, quite a few such

lexical alternatives that can be used to encode manner information. The

224 S. O zcalıskan



most typical of these alternative means is the use of adverbials that

qualify verbs of motion. It is possible to add manner to a metaphorical

motion event, such as aklımdan gecti ‘it crossed my mind’ by adding an

adverbial to the verb, as in aklımdan yel gibi gecti ‘it crossed my mind

like the wind’. Apart from adverbials, other such alternative means of ex-

pressing manner include adjectives that qualify nominalized verbs of met-

aphorical motion (e.g., his rapid fall to poverty) and verb complements

(e.g., force his mind to go blank, struggle to come out of her illness) that

convey the degree of e¤ort or force involved in a metaphorical motionevent. It is possible that Turkish speakers use these optional markings as

an alternative tool to encode components of an event type (i.e., manner)

that are not easily expressed in a simple construction in Turkish. There-

fore, as a second step, I look at the extent to which these alternative lex-

ical means are used in the two languages. The analysis has revealed the

results shown in Table 5.

As Table 5 reveals, English uses a greater amount of alternative lexical

means that encode manner than Turkish, both in novels (81 to 43 in-

stances) and newspapers (52 to 38 instances). Adverbials in both lan-guages constitute the most—and verb complements the least—frequently

used option to convey manner information. A joint look at novels and

newspapers further shows that even if both languages employ all three al-

ternative means of encoding manner, English writers use each type more

frequently than their Turkish counterparts (see Figure 4).

Some examples from the data follow (metaphorical motion events are

underlined and alternative lexical means of encoding manner are set in

contrasting type):

(27) Adverbials

a. I wanted to get out and walk eastward toward the park

through the soft twilight, but each time I tried to go I became

Table 5. Frequency distribution of alternative lexical means of encoding manner in novels

and newspapers

Novels Newspapers

English Turkish English Turkish

Adverbials 72 41 26 28

Adjectives 8 2 23 8

Verb complements 1 0 3 2

Totals 81 43 52 38




entangled in some wild, strident argument which pulled me

back, as if ropes in my chair. (Fitzgerald 1986: 36)

b. Extreme love is fed by everything. So it was that the shock of

Georgie’s decision, once the immediate pain had been

su¤ered, opened as it were a channel down which my desires

with an increased violence ran in the direction of Honor.

(Murdoch 1976: 164)

c. I felt my emotions surge painfully toward the hem of her

wrapper. (Walker 1993: 80)

d. Stocks rallied sharply for the first time in nearly two weeks . . .

(News report, Washington Post, 1 October 1999)e. If the college application process is not under way, or if it is

moving along with great ambivalence or foot dragging , it’s

time to figure out why. (Barbara F. Meltz, Boston Globe, 14

October 1999)

f. Kafamı

head-POSS:1SG-ACC

hızlı

rapid

hızlı

rapid

sallayarak

shake-CONV

bu

this

pis

rotten

du su nceyi

thought-ACC

bir

one

kovadan

bucket-ABL

copluge

waste-DAT

bosaltırcasınapour-NOM-MODIF

bosaltmak pour-INF

istedim. (Agaoglu 1998: 145)want-PAST-1SG

‘By shaking my head rapidly, I wanted to pour this rotten

thought as if I am pouring it from a bucket into the waste.’

4

98

31

2

10

68

0

10

20

30

40

50

60

70

80

90

100

adverbials adjectives verb complements

f r e q u e n c y o f u s e

EnglishTurkish

Figure 4. Frequency distribution of alternative lexical means of encoding manner (novels and

newspapers)

226 S. O zcalıskan



g. Uzak bir gecmisteki bir anıyı hatırlar . . . gibi, kameraya

bakmıslardı; karanlık

dark

bir

one

bataklıkta

swamp-LOC

agır

slowly

agır

slowly

batmaktasink-INF-LOC

olanbe-NOM

kendi their

geleceklerini future-PLU-POSS:3PL-ACC

alıskın go zlerle seyreder

watch-PRESENT

gibi, kaybettikleri

as if

belleklerinin bir daha hic geri gelmeyeceg ˘ inden kuskusu

olmayan unutkanlar gibi . . . ( Pamuk 1996: 284)

‘They stared at the camera as if they were remembering a

memory from the long forgotten past . . . ; they stared as if

they were watching their future that was sinking slowly slowly

in a dark swamp, like the forgetful people who did not havethe slightest doubt that the memory they lost would never

come back.’

h. Haftalar boyu mansetlerden inmeyen konular,

subject-PLU

birden

one-ABL

bire

one-DAT

arka

back-DAT

sayfalara

page-PLU-DAT

oradan

there-ABL

da arsivlere

archive-PLU-DAT

itiliyor

push-PASV-PRESENT

ve tozlu raflarda yıllar su recek uykusuna dalıyor. (Zulfu

Livaneli, Sabah, 15 July 1999)

‘Subjects that do not descend from the headlines for weeks, all

of a sudden are pushed to back pages and from there to the

archives, and they plunge into a years long sleep on dusty

shelves.’

i. Bo ylece iktidar sahibi olmayanların en temel

haklarının

right-PLU-POSS:3PL

sistematik

systematic

bicimde

form-LOCcig ˘ nendig ˘ i bir siddet ortamı olusur. (Aslı Erdogan, Radikal , 20

trample-PASV-NOM

July 1999)

‘As a result of this, a violent environment is created in which

the rights of those who are not in power are trampled down in

a systematic fashion.’

(28) Adjectives

a. . . . Grossman has made a steady, unspectacular climb up the

rungs of the civic leadership ladder. (David Nyhan, BostonGlobe, 25 October 1999)

b. That is, only an unexpected and monumental stumble could

undercut the solid support Mr. Bush has built among both




rank-and-file Republican voters and the party’s establishment.

(Carl P. Leubsdorf, Dallas Morning News, 29 October 1999)

c. Bu iki gelisme o ncesinde piyasanın

market-POSS

gecen hafta oldug ˘ u gibi

sıkısık

congested

seyrini

sail-POSS-ACC

koruması

protect-NOM

beklenmeli. (Esin Cetinel, Posta, 19 July 1999)

wait-PASV-NECES

‘Prior to these two developments, it must be awaited

[expected] for the market to continue its congested sailing.’

d. . . . u retimde

production-LOC

ve

and

satısta

sales-LOC

buyuk

big

bir

one

du su sfall-NOM

icinde oldug ˘ una dikkat cekiyor. (Sefer Levent,interior-LOC

Posta, 17 July 1999)

‘. . . he draws attention to being inside a big fall in production

and sales.’

(29) Verb complements

a. Thus she had evolved a kind of private commandment—those

inaudible words were simply ‘‘I must not’’—whenever the

physical female implications of her body, sexual, menstrual,

parturitional, tried to force an entry into her consciousness.

(Fowles 1981: 29)

b. On her own this year, Dole struggled to run as both the first

woman and one of the guys. (Ellen Goodman, Boston Globe,

24 October 1999)

c. I deolojinin

ideology-POSS:3SG

toplumun

society-POSS:3SG

zihinsel

mental

deg ˘ isimini

change-NOM-POSS:3SG-ACC

takip

follow-NOM

etmekte

do-NOMzorlandıgı

have.di‰culty-NOM

do nemlerde, bu teorisyenlerde de bir

period-PLU-LOC

kalite du smesi ve bag ˘ nazlasma yasaktır. (Etyen Mahcupyan,

Radikal , 21 July 1999)

‘Drop in quality and tendency towards conservatism are

forbidden among these theoreticians during periods where

ideology has di‰culty following the changes in societal

mentality.’

d. 5 þ 3 formu lu n5 þ 3 formula-POSS:3SG-ACC

o rtu lu covered-INS

birone

bicimde

form-LOC

hayata

life-DAT

gecirmeye

pass-CAUS-NOM-DAT

228 S. O zcalıskan



zorlamaktır. ( Idris Akyuz, Posta, 18 July 1999)

force-INF-PRESENT

‘He forces the 5 þ 3 formula to be passed [applied] to life in a

disguised fashion.’

The analysis at the level of alternative lexical means, including adver-

bials, adjectives, and verb complements, shows that texts in English pay

more attention to the manner dimension of metaphorical motion events

than do texts in Turkish. Even though the di¤erence is not quite as pro-

nounced as in the case of motion verbs—with Turkish writers making

quite extensive use of adjunct manner expressions—their overall use of

such expressions is nevertheless significantly lower than that of English

writers (81 to 133 instances; w2

ð1Þ ¼ 12:63, p < :001).The analysis still leaves us with an unanswered question about the

function of these alternative lexical means of encoding manner in the

two languages. Do they serve similar functions in English and Turkish,

merely acting as an alternative tool to encode manner of motion? Or is

the function di¤erent in the two languages? To answer this question we

need to look at the distribution of these adjunct manner expressions by

verb type across di¤erent metaphorical motion events. If they are more

likely to accompany manner verbs, then their function will be to augment

the manner that is already conveyed by the verb. On the other hand, if they are more likely to accompany non-manner verbs, then we may claim

that they serve as an alternative tool to add manner information to mo-

tion event descriptions that do not carry this information. Interestingly,

the results of the distributional analysis point to a di¤erence between the

two languages. As can be seen in Table 6, alternative lexical means of en-

coding manner are more likely to accompany manner verbs in English

texts (64%) and non-manner verbs in Turkish texts (72%). The di¤er-

ence suggests that English speakers use these lexical means mainly to

emphasize manner (already encoded by the verb) in their metaphorical

Table 6. Distribution of alternative lexical means (ALM) of encoding manner by verb type

in novels and newspapers

English Turkish

V: manner

þ ALM

V: non-manner

þ ALM

V: manner

þ ALM

V: non-manner

þ ALM

Novels 54 27 12 31

Newspapers 31 21 11 27

Total (percentage) 85 (64%) 48 (36%) 23 (28%) 58 (72%)




motion-event descriptions; whereas Turkish speakers use them predomi-

nantly as a means of adding manner to metaphorical descriptions that

do not carry this information at the predicate level. The rate with which

alternative manner expressions accompany manner verbs in the two lan-guages is found to di¤er significantly (64% for English, 28% for Turkish,

w2ð1Þ ¼ 14:08, p < :001).

4.1.5. Indirect evocations of manner Most of what has been discussed

so far describes the many direct ways with which one can convey manner

information in a metaphorical motion event. Of course, there are the

more indirect ways of indicating manner via descriptions of various

sorts. Here again, the pattern repeats itself, with texts in English showing

a higher incidence and more elaborate descriptions of implicit manner.

These indirect evocations include descriptions of the motion itself (exam-

ples [30], [31]), descriptions of the moving entity (32), and descriptions of

the landscape within which the motion takes place (examples [33], [34];

descriptions are underlined).

(30) For both companies and individuals, that means pacing, in one

way or another, in a race that is becoming a marathon and less a

sprint. (Steve Lohr, New York Times, 27 October 1999)

(31) So I can’t write any longer. Or only when I write fast, without

looking back at what I have written. For if I look back, then the

words swim and have no sense and I am conscious only of me,

Anna, as a pulse in a great darkness, and the words that I, Anna,

write down are nothing, or like the secretions of a caterpillar that

are forced out in ribbons to harden in the air. (Lessing 1979: 476)

(32) She was exhausted with this love for him, this physical frenzy that

was like a devil clawing and screaming inside her to get free. . . . It

took life feebly from her, tiny grasps of blood, but the devil she car-ried about with her to the most innocent places—even to church

one Sunday when she couldn’t stand the long ordeal of that day,

alone—was a creature that reached and lunged out into every part

of her body, prying, prodding, teasing, not content with anything

feeble or gentle. (Oates 1967: 191)

(33) It soothed him to indulge in these histrionics of grief, to sink to the

depths of a lurid, imponderable sadness, but even after he caught

hold of himself and began to adjust to his solitude, he never fully

recovered from Pozzo’s absence, and he went on mourning the kidas though a part of himself had been lost forever. (Auster 1990: 179)

(34) One is ejected into the world like a dirty little mummy; the roads

are slippery with blood and no one knows why it should be so.

230 S. O zcalıskan



Each one is traveling his own way and, though the earth be rotting

with good things, there is no time to pluck the fruits; the procession

scrambles toward the exit sign, and such a panic is there, such a

sweat to escape, that the weak and the helpless are trampled intothe mud and their cries are unheard. (Miller 1961: 187)

In summary, analysis of written texts at the level of motion verbs and

alternative lexical means of conveying manner shows a clear di¤erence

between English and Turkish, with English paying greater linguistic at-

tention to the manner dimension of motion events than Turkish. Turkish

writers use various alternative lexical means (adverbials, adjectives) to

add manner to their metaphorical motion descriptions; however, their

use of such manner adjuncts is still lower than that of the writers inEnglish. Texts in English also employ such alternative lexical means of

encoding manner, but mainly as a tool to strengthen the manner compo-

nent of the metaphorical motion events, already encoded by a great vari-

ety of manner verbs.

4.2. Elicitations10

4.2.1. Verbs of motion Similar to the patterns observed in written

texts, elicitations from native speakers of English and Turkish point to a

typological di¤erence in motion verb choice in metaphorical uses of spa-

tial motion. As can be seen in Figure 5, elicitations in English include

a significantly higher percentage of manner verbs (68%) than those in

7%

25%

68%

12%

46%42%

0

10

20

30

40

50

60

70

80

90

100

manner path neutral

verb type

p e r c e n t v e r b

u s e

EnglishTurkish

Figure 5. Percentage of metaphorical motion verbs in elicited responses (computed by

dividing the total number of manner, path, or neutral verbs by the total number of

motion verbs in each language)




Turkish (42%; w2ð1Þ ¼ 6:14, p < :05). The patterns hold true across dif-

ferent target domains, with English speakers consistently using higher

percentages of manner verbs than Turkish speakers (see Table 7). The dif-

ferences are particularly marked in descriptions pertaining to the domains

of time, mental and emotional states.11

The typological pattern becomes clearly evident in the diversity of the

manner verb lexicon as well. In their metaphorical descriptions, English

speakers use a manner lexicon that is almost three times as varied as that

of Turkish speakers, showing a significant di¤erence between the two lan-

guages (95 to 34 types; w2ð1Þ ¼ 28:84, p < :001). A list of all manner

verbs used by native speakers in the two languages can be found in the

Appendix.

Interestingly, the percentage of manner verbs is found to be higher in

the elicited speech than in the written texts for both English and Turkish(compare Figures 2, 3, and 5). This di¤erence becomes even more pro-

nounced in the diversity of the manner lexicon, especially for English.

English speakers use more expressive (i.e., low-frequency) manner verbs

(e.g., flitter away, e¤ervesce, shoot through, slither) in describing various

target domains in terms of spatial motion. The di¤erence is likely to be

an outcome of the task demands, with the elicitation task being construed

as a more creative exercise by native speakers. However, regardless of

the di¤erences, the typological e¤ect is highly apparent in the elicited re-

sponses, allowing a clear contrast to be drawn between the two languagesin their di¤erential attention to manner of motion. Some example adult

responses pertaining to each of the domains are presented below, in ex-

amples (35) to (43):

Table 7. Percentage of metaphorical motion verbs in elicited responses by target domain a

English Turkish

manner non-manner manner non-manner

Anger/fear 76% 24% 31% 69%

Sadness/despair 56% 44% 27% 73%

Happiness/ joy 75% 25% 47% 53%

Thoughts/feelings 76% 24% 43% 57%

Mind/consciousness 86% 14% 48% 52%

Illness/pain 58% 42% 37% 63%

Life 64% 36% 55% 45%

Death 49% 51% 37% 63%

Time 70% 30% 46% 54%

Total 68% 32% 42% 58%

a Percentages were computed by dividing the total number of manner or non-manner verbs

by the total number of motion verbs separately for each domain and each language.

232 S. O zcalıskan



(35) Time

a. Time flashes by.

b. The hours flittered away.

c. Zamantime

akıpflow-CONV

gitti.go-PAST

‘Time flew and went.’

d. Gelecek

future

elimden

hand-POSS:1SG-ABL

kacıverdi.

escape-give-PAST

‘The future escaped suddenly from my hands.’

(36) Life

a. Life drains away.

b. His life faded away.

c. Hayatlife

akıpflow-CONV

gidiyor.go-PRESENT

‘Life flows and goes.’

d. Hayat

life

gelip

come-CONV

gecer.

pass-PRESENT

‘Life comes and goes.’

(37) Death

a. Death stalks.

b. Death wafts.

c. O lu m

death

yaklasıyor.

approach-PRESENT

‘Death is approaching.’

d. O lu mler

death-PLU

birbirini

one.another-ACC

kovaladı.

chase-PAST

‘Deaths chased one another.’

(38) Sickness, pain

a. The cancer crept through him.

b. The illness bore down on him.c. Pain shoots through your limbs.

d. Kanser

cancer

midesinden

stomach-POSS:3SG-ABL

bag ˘ ırsaklarına

intestine-PLU-DAT

yu ru du .

walk-PAST

‘Cancer walked from his stomach to his intestines.’

e. Hastalık

illness

bir

one

vu cuttan

body-ABL

dig ˘ er

other

vu cuda

body-DAT

gecer.

pass-PRESENT

‘Illness passes from one body to the other.’(39) Thoughts, feelings

a. Thoughts tumble through your mind.

b. A thought popped into my head.




c. Feelings tumble through your body.

d. Du su ncelerim

idea-PLU-POSS:1SG

konudan

subject-ABL

konuya

subject-DAT

atlıyor. jump-PRESENT

‘My thoughts jump from subject to subject.’

e. Aklımdan

mind-POSS:1SG-ABL

du su nceler

idea-PLU

gecer.

pass-PRESENT

‘Thoughts pass from my mind.’

f. Duygularım

feeling-PLU-POSS:1SG

yitip

disappear-CONV

gitti.

go-PAST

‘My feelings disappeared and went away.’

(40) Minda. Mind wanders.

b. Mind drifts.

c. Aklı

mind-POSS:3SG

baska

other

yerlerde

place-PLU-LOC

dolasır.

wander-PRESENT

‘His mind wanders in other places.’

d. Gu zel

beautiful

kızı

girl-ACC

go ru nce

see-NOM

aklı

mind-POSS:3SG

cıktı.

exit-PAST

‘His mind exited when he saw the beautiful girl.’

(41) Happiness, joy

a. Happiness washed over her.

b. Happiness soars.

c. Mutluluk

happiness-PLU

yayılıyor.

spread-PRESENT‘Happiness spreads.’

d. Sevinc

joy

tasar.

overflow-PRESENT

‘Joy overflows.’

(42) Anger, fear

a. Anger pulses through one’s vein.

b. Fear crept through his bones.

c. O fkem

anger-POSS:1SG

dig ˘ er

other

insanlara

human being-PLU-DATsıcradı.

jump-PAST

‘My anger jumped to other people.’

234 S. O zcalıskan



d. I ˙cime

interior-POSS:1SG-DAT

korku

fear

du ser.

fall-PRESENT

‘Fear falls inside me.’

(43) Sadness, despaira. Sadness seeps.

b. Despairs clobbers.

c. Bu tu n

all

u zu ntu leri

sorrow-PLU-POSS:3SG

gecer.

pass-PRESENT

‘All his sorrow passes.’

d. Acı

pain

tırmanıyor.

climb.up-PRESENT

‘Pain climbs-up.’

4.2.2. Verbs of motion: Subordination of manner and aspectual mar-

king Similar to written texts, subordination is not observed in either lan-

guage, except for one instance in Turkish. The description pertained to the

domain of life ( yasam topallayarak yu ru r ‘life walks limping by’). There

are also two instances of aspectual marking, one in the domain of emo-

tional states (mutluluk kacıp gidiverdi ‘happiness escaped and went-

suddenly’), and one in the domain of time ( gelecek elimden kacıverdi

‘The future escaped-suddenly from my hand’). No instances of repeatedor aspectual verbs are observed in either of the two languages, except

for one repeated verb use in Turkish ( yasam kosar, kosar, kosar ‘life

runs, runs, runs’).

4.2.3. Alternative lexical means The elicitation task was designed to

extract only motion verbs. Nevertheless, a few participants—who an-

swered the questionnaire using full sentences—also used various alterna-

tive lexical means of conveying manner. Most of these were adverbials of manner. There is also one instance of a verb complement construction,

used by an English speaker (Time strains to move forward ). Speakers of

the two languages were comparable in their use of alternative lexical

means of encoding manner. The reason for the lack of di¤erence between

the two languages is likely to be the outcome of the Turkish speakers’

greater tendency to answer questions using full sentences, see examples

(44) to (50) (adverbials are underlined).

(44) Zamantime

cok very

hızlıfast

akıyor.flow-PRESENT

‘Time flows very rapidly.’

(45) The plague spread rapidly.




(46) Hastalık

illness

hızla

fast

ilerledi.

advance-PAST

‘Illness advanced rapidly.’

(47) His life passed quickly.(48) Hayatın

life-POSS:3SG

yavas

slow

yavas

slow

ilerledig ˘ ini

advance-NOM-ACC

farkettim.

realize-PAST

‘I realized that life is moving forward slowly slowly.’

(49) Thoughts run freely.

(50) Anger comes upon me by surprise.

In summary, the elicited responses show clear typological di¤erences atthe level of motion verbs, with English speakers using a greater number

and variety of manner verbs than Turkish speakers.

5. Concluding remarks

This article compares lexicalization options provided by a V-language

and an S-language in metaphorically extended uses of motion-event de-

scriptions. Turkish and English di¤er in their preference for encoding the

path of motion in a verb (e.g., exit, ascend ) or an associated particle (e.g.,

go out, go down), respectively. The di¤erence in how path information is

encoded has significant e¤ects on the relative codability of other semantic

components of a motion event (e.g., manner). This article focuses on the

component of manner, which is less likely to be encoded in V-languages.

The data include novels and newspapers written in English and Turkish,

along with responses elicited from adult native speakers of each language.

The article investigates the possibility that Turkish speakers may rely on

alternative lexical means to encode manner information, marking manner

outside the main verb of a clause describing a metaphorical motion event.The analysis of written texts and elicited responses from native

speakers in the two languages indicates a strong typological di¤erence

between the two language types in encoding manner of motion. Overall,

when compared to Turkish speakers and writers, English speakers and

writers encode manner of motion at a higher rate, both in the use of mo-

tion verbs (59% compared to 27%) and in using alternative lexical means

to encode it (133 instances to 81 instances; see Figure 6). English speakers

and writers also make more fine-grained distinctions within particular do-

mains of manner, using a far richer lexicon of manner verbs than Turkishspeakers/writers.

Interestingly, these alternative lexical means of encoding manner show

a di¤erence in function in the two languages. Turkish speakers/writers

236 S. O zcalıskan



use these means mainly to add manner information to their metaphorical

motion-event descriptions, thus encoding what they cannot easily convey

at the level of motion verbs. On the other hand, English speakers/writers

predominantly use these means to augment the manner that has already

been encoded by the verb in their motion event descriptions. Thus, the

same linguistic forms serve di¤erent functions in the two language types,

due to di¤erences in lexicalization patterns. However, even if Turkish

speakers/writers attempt to add manner by using various alternative lex-

ical means, the level of use still does not reach that of English speakers/

writers. Analysis of the data across written texts and adult responses

shows that sixty percent of all metaphorical motion-event descriptionsin English contains some indication of manner (i.e., a manner verb, a sub-

ordinated manner verb, or alternative lexical means of conveying manner

attached to a non-manner motion verb), whereas in Turkish, only twenty-

eight percent of metaphorical motion-event descriptions conveys some

indication of manner. In summary, the typological dichotomy proposed

by Talmy (2000) exerts itself as strongly in the metaphorically extended

uses of motion events as in the case of nonmetaphorical motion descrip-

tions, with Turkish (V-language) paying less linguistic attention to the

manner dimension of motion events, and making fewer distinctions with-in the domain.

This overall result is important in three basic respects. First of all,

it shows that cross-linguistic variation in a semantic domain—when

59%

133 instances

81 instances

27%

0

20

40

60

80

100

120

140

manner verbs alternative lexical means of manner

a m o u n t o f u s e

EnglishTurkish

Figure 6. Summary: Manner-of-motion encoding in English and Turkish (novels, newspa-

pers, elicitations; percentages were computed by averaging the total percentage of

manner verb use in each context separately in each language)




observed—is evident not only in the literal uses of the lexicon, but also

becomes apparent in the metaphorical extensions of the lexicon. In con-

trast to a literal motion event (e.g., he runs into the house) which involves

only one conceptual domain (i.e., motion in space), a metaphorical mo-tion event is composed of a source domain (i.e., motion in space), a target

domain (e.g., time, mind, states), and a conceptual mapping between the

two domains (e.g., time passage is motion along a path, mind is a con-

tainer, states are locations). Thus, the source domain of ‘‘motion in

space’’ remains the same for a literal and a metaphorical motion event,

and therefore any typological e¤ect that is evident in a literal motion

event will unavoidably be observable in the metaphorical extensions of

the event, as became evident in the data.

In fact, earlier analysis of the metaphorical structure of various concep-

tual domains in English and Turkish has revealed the structure of the

source domain to be the best candidate for language-based variation in a

metaphorical event (Ozcalıskan 2003a, 2003b). A cross-linguistic compar-

ison of metaphorical motion events in English and Turkish in terms of the

types of target domains and metaphorical mappings has shown a striking

degree of similarity between English and Turkish. Both languages are

found to conceptualize a finite, almost identical set of target domains in

terms of motion in space. Furthermore, the two languages are found to

rely on the same mappings to describe each of these target domains (e.g.,

time is a container, ideas are moving entities, seeing is touching). The

high degree of cross-linguistic similarity in the target domains and the

types of metaphorical mappings is counterbalanced by a high degree of

cross-linguistic variation in the specification of the source domain struc-

ture (Ozcalıskan 2002, 2003a). As demonstrated in this article, the lexical-

ization preferences associated with the choice of motion verbs and the use

of various other lexical means of encoding manner show a strong typo-

logical contrast between English (S-language) and Turkish (V-language).Secondly, the contrast in the encoding of manner in the source domain

is also likely to have e¤ects on our conceptualization of the target domain

of a metaphorical mapping. In a metaphorical event, the mapping is al-

ways unidirectional—from source to target—where the target domain is

conceptualized and understood in terms of the source domain. Thus, it

is the structure of the source domain that determines the structure of

the target domain, and any cross-linguistic e¤ect that is true of the source

domain—which in this case is motion events—will unavoidably be true

of the target domain. As an example, if we take the target domain of time, it is likely that English speakers have a more elaborate representa-

tion of time as a moving entity, with the ability to detect and report

more fine-grained distinctions in their experience of time (e.g., time creeps

238 S. O zcalıskan



versus time crawls versus time drags). For Turkish speakers, on the other

hand, the metaphor, time is a moving entity, may trigger a more general

sense of time passage as motion along a path (e.g., time passes), with

fewer shades of di¤erence in the way time is conceptualized to move.Thirdly, the results show that the codability of a semantic dimension in

a lexical item (e.g., a motion verb) has a spill-over e¤ect on the choice of

other lexical items or descriptions in a sentence. Since English speakers

have the option of easily encoding manner of motion in the main verb,

and use this option extensively, the manner dimension becomes a concep-

tually salient category for them. In turn, the e¤ect of this conceptual sa-

lience is reflected in the choice of other lexical items in the sentence. Thus,

English speakers encode manner not only by using manner verbs, but

also various other lexical items that encode manner (e.g., adverbials,

verb complements, indirect evocations), pointing to their greater aware-

ness of this dimension as a salient conceptual category. Relying on these

findings, it could be argued that native speakers producing texts within an

S-framed language type have recourse to mental images of metaphorical

motion-event scenes with more focus on manner of movement than do

speakers of a V-framed language.

Received 17 October 2001 University of Chicago

Revision received 22 December 2003

Appendix

List of abbreviated morphemes

ABIL abilitative marking

ABL ablative markingACC accusative marking

CAUS causative marking

CONV converb

COND conditional marking

DAT dative marking

FUTURE future-tense marking

INF infinitive marking

INS instrument marking

INTENSIFIER adjectival intensifierLOC locative marking

MODIF modifier

NEG negation marking




NOM nominalization marking

POSS possessive marking (marking varies by subject type)

PLU plural marking

PASV passive voice markingPAST past-tense marking: simple past or past progressive

PRESENT present-tense marking: generic or present progressive

1SG subject marking: first-person singular

3SG subject marking: third-person singular

1PL subject marking: first-person plural

2PL subject marking: second-person plural

List of all manner verbs observed in novels and newspapers in the two

languages

English (145 types)

bail out, blow away, bob, bolt, bounce, break away, burst, butt out, capsize,

catapult, cave into, cha-cha into, charge, chase, clamber, climb, collapse,

crash through, crawl, creep, crowd, dart, dip, dispel, drag, drain, draw,

drift, drive, ebb, elude, eject, escape, falter, filter in, flee, fleet, flit, float,

flood, flounce, flow, fly, glide, hike, hurry, inch, jack into, jerk, jump,

knock, launch, leak, leap, linger, loom out, lumber along, lunge, lurk,

march, meander, pace, percolate into, plummet, plunge, pop, port into, pour, propel, pull, pump, pursue, push, race, rally, ramble, rebound, reel,

ride, roam, roll, run, rush, rustle, sail, scramble, shoot up, shove, shower,

shuttle, sink, siphon away, skip, skyrocket, slide, slip, slop, snake, slow,

soar, somersault, spill, spin, spring, spurt, stagger, stamp out, steam

ahead /take the steam, steer, step, stream, steamroller, stride, stumble,

surf, surge, swarm, sway, sweep, swerve, swim, swing, swirl, swoop, throw,

tip out, topple, toss, track, trail, trample, tread, trip, trot, tumble, veer,

vent, verge, walk, wallow, wander, wash over/through, whirl, worm one’s

way, wrench.

Turkish (41 types)

adım at ‘step’, ak ‘flow’, at ‘throw’, atıl ‘leap’, atla ‘jump’, bat ‘sink’, cek

‘pull’, cig ˘ ne ‘trample down’, co k ‘sink’, dal ‘dive, plunge’, dolan ‘wander’,

dolas ‘wander’, do k ‘pour’, fırla ‘dash, dart’, gezin ‘stroll’, it/itele ‘push’,

kac ‘escape’, kacıs ‘run scattering’, kapıl ‘be carried away with force’, kay

‘slide’, kos ‘run’, kovala ‘chase’, sac ‘throw scatteringly’, savur ‘hurl’, sen-

dele ‘stumble’, seyir et ‘move at a steady pace’, sıcra ‘bounce’, sıyrıl ‘sneak away’, sız ‘leak’, su r ‘ride, drive’, su ru n ‘crawl’, su ru kle ‘drag’, tas

‘overflow’, takıl ‘trip’, tırman ‘climb up’, to kezle ‘stumble’, uc ‘fly’, u su s

‘crowd running’, yuvarlan ‘roll’, yu ru ‘walk’, yu z ‘swim’.

240 S. O zcalıskan



Additional verbs with the same roots, but with derivational su‰xes (14

types)

aktar ‘flow-CAUSATIVE (CAUS)’, akıt ‘flow-CAUS’, atlat ‘jump-

CAUS’, batır ‘sink-CAUS’, cektir ‘pull-CAUS’, dolastır ‘wander-CAUS’,do ku l ‘pour-CAUS’, fırlat ‘dart-CAUS’, gezdir ‘stroll-CAUS’, kacır

‘escape-CAUS’, kaydır ‘slide-CAUS’, savrul ‘hurl-CAUS’, su ru klen

‘drag-CAUS’, yu ru t ‘walk-CAUS’, and variations on verbþgive ( giriver

‘enter-give: enter suddenly’, cıkıver ‘exit-give: exit suddenly’.

List of all manner verbs used by adult native speakers in the elicitation task

English (95 types)

attack, bear down, bounce, break in, bubble up, burst (out/in), cavort,charge, climb, collapse, course through, crawl, creep, dance, drip, drag on,

drain, drift, drive, engulf, escape, escalate, evade, evaporate, fade away,

flash by, flee, flit, flitter away, float, flow, fly, gallop, hop, invade, jet, jum-

ble, jump, leap, limp, linger, lump along, lurk, march, meander, melt away,

pace, penetrate, permeate, pierce, plod by, pop, plunge, pulse, push itself,

race, rain down, rebound, retract, roll, run, rush, sag, sail, scatter, seep,

shoot through, sink, skate, skip, slide, slip (away), slither, slump, sneak

(up on), soar, spin, spring, stalk, step ahead, stomp, stream, stride, swarm,

sweep, swim, tiptoe, trickle, tumble, waft, walk, wane, wander, wash over,well up.

Turkish (34 types)

ak ‘flow’, atla ‘jump’, bas ‘step’, cag ˘ la ‘flow forcefully’, co k ‘sink’, damla

‘drip’, devril ‘collapse’, dolas ‘wander’, emekle ‘crawl’, gez ‘wander’,

hızlan ‘accelerate’, hopla ‘hop’, kac ‘escape’, kay ‘slide’, kos ‘run’, kovala

‘chase’, sacıl ‘scatter’, savrul ‘be hurled’, saplan ‘pierce’, sıcra ‘bounce’, sız

‘leak’, sıyrıl ‘squeak’, su ru n ‘creep’, takla at ‘somersault’, tas ‘overflow’,

tırman ‘climb’, topalla ‘limp,’ uc ‘fly’, yag ˘ ‘rain’, yarıs ‘race’, yavasla‘slow down’, yu ru ‘walk’, yu z ‘swim’, zıpla ‘bounce’.

Notes

* I thank Dan I. Slobin, Eve Sweetser, and Bradley M. Cooke for their helpful comments

on an earlier draft of this article, along with the Chancellor’s O‰ce, Graduate Divi-

sion, and the Center for Middle Eastern Studies the at the University of California,

Berkeley for providing research grants and fellowships that made this study possible. I

also thank Hodgen Publication Fund at the University of California, Berkeley for pro-viding a writing grant for the article. Last, but not least, I thank the two anonymous

reviewers for their helpful comments. E-mail address: [email protected].

1. The satellite-framed construction type applies to Germanic, Slavic, and Finno-Ugric

languages. Verb-framed languages include Turkic, Semitic, and Romance languages,




along with Japanese, Korean, Basque, and American and Netherlands signed

languages (Slobin 1997, 2004). In his revision of the typology, Slobin (2003) adds

another category of languages, which he calls equipollently framed languages. These in-

clude serial-verb (Niger-Congo, Hmong-Mien, Sino-Tibetan, Tai-Kadai, Mon-Khmer,

Austronesian), bipartite-verb (Algonquian, Athabaskan, Hokan, Klamath-Takelman),

and Jaminjungan languages.

2. ‘‘Manner’’ refers to a broad set of factors including motor pattern (e.g., he runs into

frustrations, the idea creeps into her mind, the day moves along at a fine tempo), or the

rate of metaphorical motion (e.g., his mind plunges in an awkward thought, she bursts

out with happiness, he suddenly comes to the realization, she slowly comes out of her ill-

ness), or the degree of e¤ort involved in the metaphorical motion (e.g., he pulls himself

out of depression, they clamber out of poverty, she struggles to enter his life, his anger

comes out violently). ‘‘Path’’ refers to the direction of motion, which, in its most elabo-

rated sense, involves metaphorical motion from a source to a goal, passing through one

or more milestones (e.g., the fear in his eyes creeps from his gaze into her heart, making its way through the labyrinths of her unconscious).

3. The typological dichotomy as presented in this article provides a general outline with-

out going into an in-depth account of some of the more recent attempts to revise the

binary typology. This is mainly because English and Turkish—as demonstrated by ear-

lier work on literal motion—constitute good exemplars of the two language types in

terms of the typological distinctions outlined by Talmy. For the interested reader, how-

ever, some of the more recent work suggests a revision of Talmy’s typological dichot-

omy, by ranking languages on a continuum of manner salience rather than assigning

them to one of the two typological categories (Slobin 2003). Some languages even

show patterns characteristic of both language types (e.g., serial-verb languages,

bipartite-verb languages), suggesting a trichotomy of language types (Slobin 2003).

4. The e¤ect of language on cognition is a highly controversial issue. Researchers seem to

be split in their findings regarding the e¤ect of language on di¤erent aspects of spatial

cognition. Some researchers argue for a clear cross-linguistic e¤ect on cognitive pro-

cesses, as demonstrated by significant performance di¤erences in various spatial navi-

gation tasks (e.g., Levinson et al. 2002), spatial relation tasks (e.g., Bowerman and

Choi 2000), and in memory tasks for motion events (Oh 2003) by speakers of di¤er-

ent languages. Some other researchers argue spatial cognition to be free of language-

based variation, as shown in memory tasks for motion events (e.g., Papafragou et al.

2002) and spatial navigation tasks (e.g., Li and Gleitman 2003). The analysis con-

ducted in this article is linguistic, and does not involve any cognitive tasks. The claimsraised about the potential e¤ects of language on cognition is only a prediction—which

is in line with the author’s position on this issue—and therefore should be regarded as

in need of future empirical verification.

5. Some of these motion-event metaphors involve more idiomatic expressions, such as he

runs for president, he runs out of time, he runs the company. These instances are also

included in the analysis, because even though these forms have become linguistically

frozen over time, they are still conceptually ‘‘alive’’. That is, they constitute conceptual

mappings that can be extended systematically and in novel ways. For instance, in the

metaphor he runs for president, the political race is conceptualized as a foot race, where

candidates correspond to athletes, and winning the presidency corresponds to crossing

the finish line in the first place. Furthermore, this metaphor can be extended in novel

ways by conceptualizing the race as a horse race (the candidates are running neck and

neck ), or a sailboat race (he is sailing smoothly towards the primaries). Similarly, in the

metaphorical expression he runs out of time, time is conceptualized as a resource and

242 S. O zcalıskan



this (linguistically frozen) metaphor can be extended systematically, as in he is short of

time, he wasted a lot of time, or he does not have enough time. Therefore, even though

the linguistic expression is conventionalized and has become frozen over time, the

mapping remains alive and can be varied systematically, leading to a variety of such

expressions that conceptualize time as resource (see Lako¤ 1987 for a more detailed

discussion of this issue).

6. The total number of motion verbs used metaphorically was 827 for the newspapers in

English, and 894 for the newspapers in Turkish.

7. The total number of motion verbs used metaphorically by adult native speakers was

380 for English and 375 for Turkish. The reason the responses did not add up to a total

of 500 verbs in either of the two languages was that subjects’ responses also included

verbs that did not express motion (e.g., time eats your life, anger boils). These instances

were excluded from the analysis, since the study only focused on metaphorical exten-

sions of spatial motion.

8. The category of manner verbs includes both bare manner verbs (e.g., run, walk, crawl )and manner verbs conflated with path (e.g., skyrocket, plummet, sink ). The category of

path verbs, on the other hand, includes only bare path verbs (i.e., verbs that only en-

code direction of motion). Lastly, the category of neutral verbs include verbs that do

not indicate either path or manner information.

9. All percentages presented in the results section are computed by dividing the number

of manner or path or neutral verbs of metaphorical motion by the total number of mo-

tion verbs used metaphorically, separately for novels and newspapers in each of the

two languages.

10. All percentages presented in the results section are computed by dividing the number of

manner or path or neutral verbs of metaphorical motion by the total number of motion

verbs used metaphorically in adult productions in each of the two languages.

11. The only two exceptions to this pattern are the domains of death and life, where the

di¤erence between the two languages in the rate of manner verb use is less marked.

However, for these two target domains, the typological di¤erence becomes more pro-

nounced when the diversity of the manner verb lexicon is considered, with English

speakers using a greater variety of manner verbs to describe these domains than Turk-

ish speakers (25 to 15 types). Yasam akar ‘life flows’ is a common metaphorical expres-

sion in Turkish that was frequently used in the elicitation task, resulting in a higher per-

centage of manner verb use in the domain of life for Turkish. Similarly, for death,

English speakers use a variety of manner verbs (e.g., drift, float, lurk, sink, slide, stalk,

waft) in describing death as a moving entity, while Turkish speakers use only a fewtypes of manner verbs (e.g., kovala ‘chase’, yu ru ‘walk’).

References

Bowerman, Melissa and Sonja Choi

2000 Shaping meanings for language: Universal and language specific in the ac-

quisition of spatial semantic categories. In Bowerman, M. and S. C. Levin-

son (eds.), Language Acquisition and Conceptual Development. Cambridge:

Cambridge University Press, 475–511.Ibarretxe-Antunano, Iraide

2001 Motion events in Basque narratives. Paper presented at the 7th International

Conference on Cognitive Linguistics, Santa Barbara, CA.




Lako¤, George

1987 The death of a dead metaphor. Metaphor and Symbolic Activity 2 (2), 143–

147.

Lako¤, George and Mark Johnson

1980 Metaphors We Live By. Chicago: University of Chicago Press.

1999 Philosophy in the Flesh. New York: Basic Books.

Levinson, Stephen C., Sotaro Kita, Daniel B. M. Haun, and Bjorn H. Rasch

2002 Returning the tables: Language a¤ects spatial reasoning. Cognition 84, 155–

188.

Li, Peggy and Leila Gleitman

2002 Turning the tables: Language and spatial reasoning. Cognition 83, 265–

294.

Naigles, Letitia, A. Eisenberg, E. Kako, M. Highter, and N. McGraw

1998 Speaking of motion: Verb use by English and Spanish speakers. Language

and Cognitive Processes 13 (5), 521–549.Oh, Kyung-ju

2003 Language, cognition and development: Motion events in English and

Korean. Doctoral dissertation in Psychology, University of California at

Berkeley.

Ohara, Kyoke H.

1999 Cognitive and structural constraints in motion event descriptions: Ob-

servations from Japanese and English. Poster presented at the 16th Annual

Meeting of the Japanese Cognitive Science Society Joint Conference. Tokyo,

Japan.

Ozcalıskan, Seyda

2000 Contrastive e¤ect of narrative perspective vs. typological constraints in

encoding manner of motion. Poster presented at the 7 th International Prag-

matics Conference, Budapest, Hungary.

2002 Metaphors we move by: A crosslinguistic-developmental analysis of meta-

phorical motion events in English and Turkish. Ph.D. dissertation in Psy-

chology, University of California, Berkeley.

2003a Metaphorical motion in crosslinguistic perspective: A comparison of English

and Turkish. Metaphor and Symbol 18 (3), 189–228.

2003b In a caravanserai with two doors, I am walking day and night: Metaphors of

death and life in Turkish. Cognitive Linguistics 14 (4), 282–320.

2004 Encoding the manner, path, ground components of a metaphorical motionevent. Annual Review of Cognitive Linguistics 2, 73–102.

Ozcalıskan, Seyda and Dan I. Slobin

1999a Learning ‘‘how to search for the frog’’: Expression of manner of motion in

English, Spanish, and Turkish. In Greenhill, A., H. Littlefield, and C. Tano

(eds.), Proceedings of the 23rd Annual Boston University Conference on Lan-

guage Development. Somerville, MA: Cascadilla Press, 541–552.

1999b How children encode motion events in two types of languages. Poster pre-

sented at IASCL-Eight International Congress for the Study of Child Lan-

guage, San Sebastian, Spain.

2000a Climb up vs. ascend climbing: Lexicalization choices in expressing motion

events with manner and path components. In Catherine-Howell, S., S. A.

Fish, and Keith Lucas (eds.), Proceedings of the 24th Annual Boston Univer-

sity Conference on Language Development. Somerville, MA: Cascadilla

Press, 558–570.

244 S. O zcalıskan



2000b Expression of manner of movement in monolingual and bilingual adult

narratives: Turkish vs. English. In Goksel, Aslı and Celia Kerslake (eds.),

Studies on Turkish and Turkic Languages. Wiesbaden: Harrassowitz Verlag,

253–262.

2003 Codability e¤ects on the expression of manner of motion in English and

Turkish. In Ozsoy, A. S. D. Akar, M. Nakipoglu-Demiralp, E. E. Taylan,

A. Aksu-Koc (eds.), Studies in Turkish Linguistics. Istanbul: Bogazici Uni-

versity Press.

Papafragou, Anna, Christine Massey, and Lila Gleitman

2002 Shake, rattle, ‘n’ roll: The representation of motion in language and cogni-

tion. Cognition 84, 189–219.

Slobin, Dan I.

1996 Two ways to travel: Verbs of motion in English and Spanish. In Shibatani,

M. and S. A. Thompson (eds.), Essays in Semantics. Oxford: Oxford Uni-

versity Press, 195–317.1997 Mind, code, and text. In Bybee, J., J. Haiman, and S. A. Thompson (eds.),

Essays on Language Function and Language Type: Dedicated to T. Givo n.

Amsterdam: John Benjamins, 437–467.

2000 Verbalized events: A dynamic approach to linguistic relativity and determin-

ism. In Niemeier, S. and R. Dirven (eds.), Evidence for Linguistic Relativity.

Amsterdam: John Benjamins, 107–138.

2003 Language and thought online: Cognitive consequences of linguistic relativ-

ity. In Gentner, D. and S. Goldin-Meadow (eds.), Language in Mind: Ad-

vances in the Investigation of Language and Thought. Cambridge, MA: MIT

Press, 157–191.

2004 The many ways to search for a frog: Linguistic typology and the expression

of motion events. In Stromqvist, S. and L. Verhoeven (eds.), Relating Events

in Narrative: Typological and Contextual Perspectives. Mahwah, NJ: Law-

rence Erlbaum Associates, 219–257.

Talmy, Leonard

1985 Lexicalization patterns: Semantic structure in lexical forms. In Shopen, T.

(ed.), Language Typology and Semantic Description, vol. 3: Grammatical

Categories and the Lexicon. Cambridge: Cambridge University Press, 36–

149.

2000 Toward a Cognitive Semantics. Cambridge, MA: MIT Press.

Literary texts

Agaoglu, Adalet

1998 Bir du g ˘ u n gecesi [The night of a wedding]. Istanbul: Yapı Kredi Yayınları.

Aral, Inci

1999 I ˙cimden kuslar go cu yor. [Birds migrate from my inside]. Istanbul: Can

Yayınları.

Auster, Paul

1990 The Music of Chance. New York: Viking Press.

Bowles, Paul

1966 Up and Above the World . New York: Simon and Schuster.

Cumalı, Necati

1998 Ask da gezer [Love also wanders]. Istanbul: Cagdas Yayınları.




Fitzgerald, F. Scott

1986 The Great Gatsby. New York: Charles Scribner’s Sons.

Fowles, John

1981 The French Lieutenant’s Woman. New York: The New American Library.

Kemal, Yasar

1997 Al Go zu m Seyreyle Salih [Salih, take my eye and watch]. Istanbul: Adam

Yayınları.

Lawrence, D. H.

1980 Lady Chatterley’s Lover. New York: Bantam Books.

Lessing, Doris

1979 The Golden Notebook . New York: Bantam Books.

Miller, Henry

1961 The Tropic of Cancer. New York: Grove Press.

Mungan, Murathan

1993 Son I stanbul [The last Istanbul]. Istanbul: Metis Yayınları.Murdoch, Iris

1976 A Severed Head . London: Penguin.

Oates, Joyce Carol

1967 A Garden of Earthly Delights. New York: The Vanguard Press.

Ozlu, Tezer

1994 Cocuklug ˘ umun sog ˘ uk geceleri [The cold nights of my childhood] Istanbul:

Yapı Kredi Yayınları.

Pamuk, Orhan

1996 Kara Kitap [The black book]. Istanbul: Iletisim Yayınları.

Soysal, Sevgi

1996 Yenisehir’de bir o g ˘ le vakti [An afternoon in Yenisehir]. Ankara: Bilgi Yayı-

nevi.

Uzuner, Buket

1997 Kumral Ada Mavi Tuna [Brown Ada, blue Tuna]. Istanbul: Remzi Kitabevi.

Walker, Alice

1993 Possessing the Secret of Joy. New York: Pocket Star Books.

Yeginobalı, Nihal

1997 Sitem [Sitem]. Istanbul: Can Yayınları.

Newspapers

Boston Globe Posta

Cumhuriyet Radikal

Dallas Morning Star Sabah

Hu rriyet The New York Times

Los Angeles Times Washington Post.

246 S. O zcalıskan

Date post:	06-Apr-2018
Category:	Documents
Upload:	agnieszka-hamann
View:	217 times
Download:	0 times

Cogl.2005.16.1 Ozacaliskan Motion Events Metaphor

Documents