Two types of Intransitive Vs: The Unaccusative Hypothesis FRAMEWORK - - PowerPoint PPT Presentation

1

Learner corpora and the acquisition of word order:

A study of the production of Verb-Subject structures in L2 English CORPUS LINGUISTICS 2007

Birmingham

Cristobal Lozano

Universidad de Granada

Amaya Mendikoetxea,

Universidad Autónoma de Madrid/Lancaster University

http://www.uam.es/woslac

2

AIM OF THE PRESENTATION

To inform on the results of a study on the production of

postverbal subjects (VS order) in non-native English.

Purpose of the study: to characterize the interlanguage

• f advanced non–native speakers of English L2

(Spanish/Italian L1) by examining their production of

both grammatical and ungrammatical VS structures, as represented in the relevant ICLE subcorpora

(Granger et al. 2002), :

3

FRAMEWORK

Main question:

What are the conditions under which learners produce inverted subjects (VS structures), regardless of problems to do with syntactic encoding (grammaticality)?

Comparative Framework: to determine the role of L1 in

L2 acquisition (transfer) in the areas under study

Learner corpora vs. native corpora (LOCNESS) ENGLISH and SPANISH/ITALIAN differ in devices employed for constituent ordering: English ‘fixed’ order is determined by lexico-syntactic properties and Spanish/Italian ‘free’ order is determined by information structure, syntax-discourse properties.

4

Two types of Intransitive Vs: The Unaccusative Hypothesis

Unaccusative V : S is a notional object (patient or

theme) - an entity that comes into existence (3), appears on the scene (4) or undergoes a change of state/location (5)

(3) Problems exist (4) Three girls arrived (5) The window broke

Unergative V: S is a notional subject; it is an agent

• r has protagonist control over the action:

(6) John spoke/cried/laughed/worked…

John is a subject both notionally and syntactically

5

Word Order in L1 English (1)

Fixed SV(O) order- Restricted use of postverbal subjects:

a)

XP V S (Inversion structures with an opening adverbial) (7) Michael puts loose papers like class outlines in the large file-size pocket. He keeps his checkbook handy in one of the three compact pockets. The six pen and pencil pockets are always full and <in the outside pocket> go <his schedule book, chap stick, gum, contact lens solution and hair brush>. [Land’s End March 1989 catalog. p. 95] (Birner 1994: 254) (i) XP is an adverbial element, typically expressing time or place and linking the sentence to the prior discourse (ii) V is an intransitive verb, typically expressing existence or appearance on the scene (= unaccusative) (iii) S is often syntactically/phonologically ‘heavy’ consisting of a noun and a variety of pre and/or postmodifiers, which introduce new information in the discourse.

6

Word Order in L1 English (2)

b) There-constructions

(8)

• a. Somewhere deep inside [there] arose a desperate hope that he

would embrace her [FICT ]

• b. In all such relations [there] exists a set of mutual obligations in

the instrumental and economic fields [ACAD]

• c. [There] came a roar of pure delight as…. [FICT]

[Biber et al. 1999: 945]

• Roughly the same (sub)class of Vs and same conditions as in XPVS

structures.

7

Word order in L1 English (VS order)

• Lexicon-syntax interface (Levin & Rappaport-Hovav, etc):
• Unaccusative Hypothesis (Burzio 1986, etc)

!

• Syntax-discourse interface (Biber et al, Birner 1994, etc):
• Postverbal material tends to be focus/relatively unfamiliar information

"#$%#

• Syntax-Phonological Form (PF) interface (Arnold et al 2000, etc)
• Heavy material is sentence-final (Principle of End-Weight, Quirk et al.

1972) – general processing mechanism (reducing processing burden)

&'#() *) ## # +,-./0

Subjects which are focus, long and complex tend to occur postverbally in those structures which allow them (unaccusative Vs).

8

Word Order L1 Spanish/Italian (1)

Postverbal subjects are produced ‘freely’ with all verb classes:

(13) a. Ha telefoneado María al presidente. (transitive). Has phoned Mary the president

• b. Ha hablado Juan.

(unergative) c. Ha llegado Juan (unaccusative) has spoken Juan . has arrived Juan

• Inversion as ‘focalisation’:
• preverbal subjects are topics (given information)
• and postverbal subjects are focus (new information) (Belletti 2001, 2004,

Zubizarreta 1998)

(14) ¿Quién ha llegado/hablado? (15) Chi è arrivato/parlato? Who has arrived/spoken? i. Ha llegado/hablado Juan

• i. É arrivato/ A parlato Gianni

ii. #Juan ha llegado/hablado

• ii. # Gianni é arrivato/a parlato

9

Word Order L1 Spanish/Italian (VS order)

Lexicon-syntax interface

1/ 2(

Syntax-discourse interface

0 2'

Syntax-Phonological Form (PF) interface Heavy subjects show a tendency to be postposed – a universal language processing mechanism: placing complex elements at the end reduces the processing burden (J. Hawkins 1994).

Subjects which are focus, long and complex tend to occur postverbally, with no restrictions at the lexicon-syntax interface.

10

The phenomenon in SLA

Production of postverbal subjects in L2 English (Zobl 1989 Rutherford 1989, Oshita 2004)

Only with unaccusative verbs (never with unergatives).

• Unaccusatives: arrive, happen, exist, come, appear, live…
• Unergatives: cry, speak, sing, walk ...
• L1 Spanish/Italian/Arabic – L2 English:

3 4% 5##6 ( (#

• Explanation: syntax-lexicon interface (Unaccusative Hypothesis)

11

The psychological reality of the Unaccusative Hypothesis

A number of studies have found that L2 learners are

aware of the argument structure distinction between unaccusative and unergative Vs and that they use this as a guiding principle to construct L2 mental grammars.

However, learners have difficulty in determining the

range of appropriate syntactic realizations of the distinction, and this can persist into near-native levels of proficiency (see R. Hawkins 2001: 5.4).

CRUCIAL DIFFERENCE: these previous studies focused on ERRORS, thus emphasising the differences between native and non-native structures. By contrast, our study emphasises the similarities between native and non-native structures.

12

Hypotheses

GENERAL HYPOTHESIS:

Conditions licensing VS in L2 Eng are the same as those in

Native Eng, DESPITE differences in syntactic encoding. SPECIFIC HYPOTHESES:

:

Postverbal subjects with unaccusatives (never with unergatives)

!:

Postverbal subjects: heavy (but preverbal light)

"!#$%:

Postverbal subjects: focus (but preverbal topic)

13

Method

• Learner corpus: L1 Spa – L2 Eng; L1 Ital – L2 Eng
• ICLE (Granger et al. 2002)

(Problem: proficiency level?)

• WordSmith v. 4.0 (Scott 2004)
• Concordance queries can be performed automatically with WordSmith, by

targetting specific verbs BUT there is a lot of manual work (filtering out unusable data, coding data in Excel, analysing data in SPSS, etc).

• SPSS v. 12.0

Corpus Number of essays Number of words ICLE Spanish 251 200,376 ICLE Italian 392 227,085 TOTAL 643 427,461

14

• !"
• #$

!%

• #$

# %$ !% & 6

• #'
• (
• .78+9''&1
• (
• '&:1;+9''&1
• $

(( !

• #
• ($
• $$
• $
• '9+..+9'
• (
• (
• ':<'%1,+

+9''&1 (

• '##

#

• 9%11+=&>

'0+%?17

• )(( !
• %.?@+=<'

$

• $
• <=+%8+@+=<'
• * ! $+

)! )#''

• %')$+(!'
• #
• (

DATA ANALYSIS

• Based on Levin (1993) and Levin & Rappaport-Hovav (1995):
• Unergatives: cough, cry, shout, speak, walk, dance…
• [TOTAL: 41]
• Unaccusatives: exist, live, appear, emerge, happen, arrive…
• [TOTAL: 32]

15

WordSmith: query searches:

For every lemma (e.g., APPEAR, ARISE), we

searched for:

All possible native forms:

• appear, appears, appearing, appeared
• arise, arises, arising, arose, arisen

All posible overregularised and overgeneralised learner

forms:

• arised, arosed,arisened, arosened (“So arised the Sain

Inquisition”)

All possible forms with probable L1 transfer of spelling:

• apear, apears, apearing, apeared

All other possible misspelled forms:

• appeard, apeard

16

Data analysis (cont’d)

• CONCORDANCES: RAW OUTPUT
• Thousands of concordances, BUT approx. ¾ were unusable.
• Filtering criteria had to be applied manually.

17

Data coding/analysis: EXCEL

18

H1: syntax-lexicon

!!A !A A 5!A !!A !A 4A 3!A

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% SV VS SV VS Unerg Unac % of production Spanish Italian

Subcorpus V type # postverbal S # usable concordances Rate (%) Spanish Unergative 153 0/153 (0%) Unaccusative 52 640 52/640 (8.1%) Italian Unergative 143 0/143 (0%) Unaccusative 15 574 15/574 (2.6%)

19

Examples: syntax-lexicon (unaccusatives only)

• ,#-,./012/314

506 (#$2 6

• ,%-,.712814
• 6

, # ' 9

• : ;

: :

20

Results: Unaccusative: grammatical vs. ungrammatical VS

Figure 1. Proportion (in %) of grammatical vs. ungrammatical unaccusative VS

65.4% 34.6% Ungrammatical Grammatical Grammaticality Group: VS Spanish ICLE 53 . 3% 46 . 7% Ungrammatical Grammatical Group: VS Italian ICLE

21

H2 : syntax-phonology

1 2 3 4 5 6 7 8 9 1011 1213141516171819202122 232425

Weight (# of words)

VS Spanish ICLE VS Italian ICLE SV Spanish ICLE SV Italian ICLE

Group

• 22

Examples: syntax-phonology

SV: typically LIGHT

B ##( # 6-

VS: typically HEAVY

C%#'< % = 6#'

• 23

Result 3: syntax-discourse

A 5A !3A CA !!A !!!A A 55A

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Top Foc Top Foc SV VS Spanish Italian

Discourse status (topic/focus) has to be measured manually by establishing theoretical criteria and then by checking the context (or even the essay) manually

24

EXAMPLES: syntax-discourse

VS: FOCUS

D6 ' 9 55!>

• SV: typically TOPIC

3 (( ##(( # (/ ( $ # (# 6

25

• 15. Summary/Conclusion

&% %'(

Syntax-discourse………………………………… FOCUS Syntax-phonology……………………………………. HEAVY

%'( &%

Syntax-discourse…….. TOPIC Syntax-phonology……… LIGHT

& &

26

Our results and CIA (1)

• Contrastive Interlanguage Analysis (Granger 1996, Gilquin 2001)

(a) NNS vs. NNS: different non-native data. By comparing learner data from different L1 backgrounds, we can gain a better understanding of interlanguage processes and features, such as those which are the result

• f transfer or those which are developmental, common to

learners with different L1.

No significant differences between Italian and Spanish learners, as expected, except for frequency of inversion [8.1 (Sp) vs. 2.6 (It))]

Possible explanation: lexical bias.

27

Result: VS and specific unaccusative verbs

28

Our results and CIA (2)

(b) NNS vs. NS: non-native vs. native data.

It involves a detailed analysis of linguistic features in native and non-native corpora to uncover and study non-native features in the speech and writing of (advanced) non-native speakers. This includes errors, but it is conceptually wider as it seeks to identify

• veruse and underuse of certain linguistic features and patterns

(Granger 2002: 12-13).

Corpus Number of essays Number of words ICLE Spanish 251 200,376 UAM 85 63,836 TOTAL 336 264,212

• vs. LOCNESS:

436 324, 304

29

Syntax-Lexicon

NS vs. NNS

100,0% 0,0% 92,9% 7,1% 100,0% 0,0% 97,8% 2,3% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% SV VS SV VS Unerg Unac Spanish ICLE+WriCLE LOCNESS

Percentage of subjects produced (group x verb type)

30

Examples of VS in LOCNESS

AdvP insertion (3 tokens)

(27) Thus began the campaign to educate the public

• n how one contracts Aids.

PP insertion (7 tokens)

(28) …along with this development has come opposition from both

the medical field and a proportion of the general public, who…

There-insertion (6 tokens)

(29) Certainly there exists a demand for this work to be done.

31

Syntax-Phonology: NS vs. NNS

Heavy-Light subjects with Unacc Vs

• Subjects in SV structures with unacc Vs tend to be ‘light’ in NS

(67.7% in ICLE+WrICLE and 68.1% in LOCNESS )

• Subjects in VS structures with unacc Vs are overwhelmingly ‘heavy’

in NS (81.0% in ICLE+WriCLE and 81.3% in LOCNESS).

Weight_nom

Porcentaje

VS Locness VS Spanish WriCLE+ICLE Group

32

Syntax-Discourse: NS vs. NNS

Focus and Topic subjects with Unacc Vs

• Most subjects in SV structures with unacc Vs in NS are topic (89.9%

in ICLE+WriCLE and 83.5% in LOCNESS); just a few are focus (10.5% in ICLE+WriCLE and 16.5% in LOCNESS; p=0.223).

• Subjects in VS structures with unacc Vs are overwhelmingly focus in

NS (98.3% in ICLE+WriCLE 100% in LOCNESS; p=0.784).

Discourse_postverbal

Porcentaje

VS Locness VS Spanish WriCLE+ICLE Group

33

NNS vs. NS: comparisons across different NNS

97,4 91,9 92,9 97,7 97,8 2,6 8,1 7,1 2,3 2,2 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Italian ICLE Spanish ICLE Spanish ICLE & WriCLE French ICLE LOCNESS Frequency (%) of VS production SV VS

34

Ns: Verbs in VS structures

LOCNESS: Inv/Totalinvs

0,0 % 0,0 % 0,0 % 0,0 % 6,3 % 56,3 % 0,0 % 0,0 % 0,0 % 0,0 % 0,0 % 0,0 % 0,0 % 31,3 % 0,0 % 0,0 % 0,0 % 0,0 % 0,0 % 0,0 % 0,0 % 0,0 % 0,0 % 0,0 % 0,0 % 6,3 % 0,0 % 0,0 % 0,0 % 0,0 % 0,0 % 0,0 % 10,0 % 20,0 % 30,0 % 40,0 % 50,0 % 60,0 %

%00+%= %==@+ <+71 ;+@+.&0 ;'%00+%= +9+=7+ +',%0+ >%.. >&..&" 7=&" 8;+ .@+ 0%'' =+:=1 '+.+ ':=@@+

35

Some preliminary conclusions

These results confirm that Spanish (and, presumably, Italian)

learners of English produce postverbal subjects under exactly the same interface conditions as in L1 English (unaccusativity being a necessary but not a sufficient condition).

Spanish and Italian learners show persistent problems in the

syntactic encoding of the construction, producing mostly ungrammatical examples (it-insertion, 0-insertion, wrong XP).

Spanish learners overuse the construction and show a lexical

bias for the V exist.

36

Thank you!

37

! ? ! 5@A

• 2B 2<2$2!" 50666A2C*,>D223E,07-FF
• %25066@A2C GH ID2,*;JB-+(;5A2E6-/6
• %25066KA25A 8+;,'>(
• %25066KA2(2,$!II 5A 0+;,'>(
• % 2)2B2"$2 ? 5@///A2$,$
• %2%5@//KA2 2 36,088-0F/
• %2%5@//FA2( 2 /3,008-0FE
• %2)5@/33A2 !" $,$
• %2B5@//KA2$"2 36,30-@8@
• %I2$5@/7EA2#" ),L
• 2"2$I2!BMI2#;>#5A2C!2-?,ND25A2O

#

• ;25@/7@A2$"),?
• ;25@//FA2!22#,#(
• 2(<!)$5066@A2 , 2% @/20,078-8@6
• )PI2$5066KA2! &# %,()
• 2$'?QI 5066KA'##,"
• ?QI-2'5@/7/A2C-N((D2!" @@,007-08/
• ?QI-2'5@//8A2CRD2 ()*@@,@@8-@F@
• ?2B5@//0A2 $",(
• ?I2$5066KA2C5)A?D2+,-,2

(2'066K

• ";2!5@//@A2C ,-2, 5A2 &./"

%,#"2@K@-@E3

• "92"5066@A2C#D2 85@A,/F-@08
• "25@//EA2C?;,ID2,LN2% %#B #5A2

#01022$,$(283-F@

• "250660A2CS-D2,"2B*(-5A2".-

$$$EJ(,%N28-88

• "2 )> ?# 50660A2)1 3045!$--,($
• *;2B5@//KA2.5,(
• *2B50668A2C$>9 D242@/,038-86K
• *2-B5@/7KA2C'D2-2@F,F8@-F3K
• *;22B-+(;5A5066@A2 16$4/$'>>'>,'>(
• B2'L 5A5@/7/A2%16),L
• LT;2"5066KA20#0#&.#+#<,%U
• LT;2"5066FA2C-> ,D272@650A,@@/-@F/
• LI;2(5066@A2C ,% D2,(!2<2# 2* 5A2899:

$,!$28@0-800

• L2"5@//7A2.$J+;,$
• L2*J) 5@//@A2CND2 7F,0@@-0F7
• $2%5@//8A2;1.&.$,(
• $2%#!-*5@//FA2/$#0#22#,#(
• $2B2% 5@//KAC$,QP9ORD2) F,K8-77

38

! ? ! 50A

• $I250668A/$&-+0$ >,()
• $I25066EAC?,9-D24 00,@-K8
• $I2#;>50663A2C(N$0 ,D2"92"2%)PI ( 5A

$4!

• $I2#;>5A2CN,CD -D25A2"V

O #

• $NQ2#5@///AC >OOD,%9 ) 5A2*$<2#, -2@03F-@8@F
• (2)5@/37A2CD2.!2K,@F3-@7/
• '2*50666A2C<,CD $0 D24 @E,0/8-80K
• '2*50660A2C,N-$0 D2 @,KF-E@B$
• '2*5066KA2C W>D24 06,/F-@86
• (#PI J#PI- 5066EA
• (25@//7A2C' >(N(>ND21" @6,0@/-0K@
• (2#5@//3A2$162,$
• (2 ?5@/7@A2C>"-D ,(5A242$,(2008-0FF
• (2 ?5@//0A2CX(",N2D2,2<#25A233&3$. 4#2

,B%N20/F-80F

• Y;2!2"2"$JN; 5@/30A2 .$" $,$
• !II $5@/70A2# ),?
• !II2$5@//3A2CD2,$*5A2)1$# @),L (
• !II2$5066KA2C$D2,% 5A2008-0F@
• !2#5@/7EA2 $2,B%N
• !2<5@/7EAC"$09,D2 4" 0,@-@F
• !2<5@/7/ACD2," B
• 5A2$.-2,(2@E8-@70
• 2L5@/7FA22,(
• 2#50660A25# =$>9?" '>,'>(
• 25@//8A2CD24 /,00-K3
• 25@//FA2C9;D2,$ ;2$;2#-5A2&

) 4,B%N2@F8-@3F

• P2 5@//6A22()2(
• <25@//3A2 -;D274"0E,8K3-8E@
• <25@//3AC -<;D(D7420E58A,8K3-8E@
• <2J2B50668AC( D!2"%# 5A3;,

(

• +250660A2C$09D2,B#B? 5A..899:.-$,

Z ( $207/-0/E

• X2L50660A2#2,(
• X2*5@/7/AC $09D2," B 5A2$.-2

,(2068-00@

• XI2#$5@//7A2" 522#,#(
• XI2#$5@///AC$,D2%9 ) 5A2*$<=$@?2#,

39

ADDITIONAL SLIDES TO FOLLOW:

40

Unaccusativity Hypothesis

(1)

• a. unergative
• b. unaccusative

‘John spoke’ ‘Three girls arrived’

41

ST

• SD

T’ pro T SV llegó

• V

SD llegó un hombre

VS in native Spanish

)*%+ ,-./0 Inergativos: SV Inacusativos: VS

A: ¿Qué pasó anoche en la reunión? B: Un hombre gritó. # Gritó un hombre. A: ¿Qué pasó anoche en la reunión? B: # Un hombre llegó. Llegó un hombre.

ST

• SD

T’ un hombre T SV gritó

• SD

V un hombre gritó

42

ST

• SD

T’ pro T SFoc llegó

• SD

Foc’ un hombre

• [+Foc] Foc

SV [+Foc] V SD llegó un hombre [+Foc] ST

• SD

T’ pro T SFoc gritó

• SD

Foc’ un hombre

• [+Foc] Foc

SV [+Foc] SD V un hombre gritó [+Foc]

VS in native Spanish (2)

!). ,-1%0 Inergativos: VS Inacusativos: VS

A: ¿Quién gritó anoche en la reunión? B: # Un hombre gritó. Gritó un hombre. A: ¿Quién llegó anoche a la reunión? B: # Un hombre llegó. Llegó un hombre. [-interp] [+interp] [+interp] [-interp]

43

Data analysis (cont’d)--------

• CONCORDANCES: 6 BASIC FILTERING CRITERIA:

The verb must be intransitive (unergative or unaccusative).

• In the screen of the television one or two “rombos” should appear. [unac]
• Leontes cries and the statue talks. [unerg]
• This government’s movement has created several opinions. [trans]

The verb must be finite, with(out) aux.

• …also it exists the psychological agresssions… [finite no aux]
• … the cases of men mistreated do not appear in the media. [finite aux]
• This contradiction could disappear [finite modal]
• There’s no reason for it to exist. [for clause + to inf]
• Poor people cross borders to escape from poverty. [to-inf clause]
• …let time pass… [‘let’ constructions]
• …make everyone’s life go ahead [causative + infinitive]
• Returning to the title of this paper,… [gerundive clauses]
• …they go away in order to escape to France. [‘in order to’ clauses]
• …women have to live with the agressor [have to/ought to/able to]
• …prudence was beginning to disappear. [verbal/aspectual periphrases]
• Before entering the argumentation,… [small clauses]
• …instead of following… [complement of P]
• …likely to happen… [complement of A]
• The tests to enter the army are quite difficult now. [complement of N]

44

Data analysis (cont’d)--------

The verb must be in the active voice.

• This contradiction could disappear. [active unaccusative]
• This situation has already been happened. [passivised unaccusative]

The subject must be an NP.

• …it arose [diverse social ranks, the rich and the poor that depended on the

property they had]. [inverted NP subject]

• …it only remains [to add that nowadays we live in a world…] [extraposition]
• It happened [that the countries which make the weapons are…] [extraposition]

The sentence can be either grammatical or ungrammatical in native English.

• This contradiction could disappear. [gram]
• …it won’t exist nothing of what people don’t get bored or tired. [ungram]

The subject can appear either postverbally (VS) or preverbally (SV).

• …the real problem appears when they have to look for their first job. [SV]
• So arised the Saint Inquisition. [VS]

45

Data analysis (cont’d) ---------

• OTHER FILTERING CRITERIA
• Target V + V (verbal coordination)
• Families without father exist and work well.
• Coordinator + target V
• …we can manage to obtain it and live in a better world.
• Interrogatives (only if V is the target)
• How could they live?
• Does exist then a manipulation of television?
• Formulaic & Set expressions in English
• As sometimes happens…
• …fall victim to…
• …the world we live in.
• Set expressions transferred from the L1
• …it happens the same.
• …they fall into account that they have treated very badly Mr Hardcastle.
• Phrasal verbs:
• …a scientist come up with an intention…
• Quotes (literary or other):
• “To what purpose, April, do you return again?
• “Feminism has to evolved or die”, Friedan said in 1982…

46

WordSmith: query searches:

For every lemma (e.g., APPEAR, ARISE), we

searched for:

All possible native forms:

• appear, appears, appearing, appeared
• arise, arises, arising, arose, arisen

All posible overregularised and overgeneralised learner

forms:

• arised, arosed,arisened, arosened (“So arised the Sain

Inquisition”)

All possible forms with probable L1 transfer of spelling:

• apear, apears, apearing, apeared

All other possible misspelled forms:

• appeard, apeard

47

Extraposition was discarded

NOTE: extraposition discarded:

#=4 ;=4

48 there-insertion AdvP-insertion Ø-insertion XP-insertion Loc inversion it-insertion 100% 95% 90% 85% 80% 75% 70% 65% 60% 55% 50% 45% 40% 35% 30% 25% 20% 15% 10% 5% 0%

Production rate (%)

13 20 7 33 27 12 10 10 15 15 38

VS Italian ICLE VS Spanish ICLE

Group

Result: Unaccusative: Type of VS structures

49

Result: VS and (in)definiteness

Figure 1: Production of postverbal subjects according to their definiteness. Definite 41.4% Indefinite 58.6%

DEFINITE

• #
• ,

# INDEFINITE

• ,

#(#(

• ;=

50

Scale (syntactic weight/complexity)

NOMINAL SCALE ORDINAL SCALE SYNTACTIC STRUCTURE (D) N PRN LIGHT 1 (D) ADJ N (D) N (D) ADJ* N 2 (D) (ADJ) N* PP (D) (D) (ADJ) N N PP* AdjP* (D) ADJ N PP (D) N IP/CP (D) (ADJ) N* PP* HEAVY 3 (D) ADJ N* (PP*)

51

Topic vs. Focus: Retrievability scale

$%.1, 0 $%2,!#0

• !

! ! !

• !!

" " " "#! #! #! #!

• $

$ $ $ !!

• %

% % %& & & & !