Transcript Powerpoint slides - Lancaster University

Passive constructions in English and
Chinese: a corpus-based study
Tony McEnery
Richard Xiao
1
16/07/2005
CL 2005, Birmingham
Aims and objectives

Using comparable corpus data
–
–
–
2
to explore passives in written and spoken English
to explore passives in written and spoken
Chinese
to contrast passives in the two languages
16/07/2005
CL 2005, Birmingham
Corpora

English
–
–

FLOB: ca. one million words, written British English, 500
samples, 15 text categories, 1991-1992
BNCdemo: ca. four million words, the demographically
sampled component of the BNC (conversational data)
Chinese
–
LCMC: ca. one million words, written Mandarin Chinese,
500 samples, 15 text categories, 1991-1992

–
3
http://www.ling.lancs.ac.uk/corplang/lcmc
LDC CallHome Mandarin: ca. 300,000 words, telephone
conversations, 120 transcripts of 5-10-minute continuous
telephone conversations
16/07/2005
CL 2005, Birmingham
Text categories covered FLOB/LCMC
4
Code
Text category
No. of samples
Proportion
A
Press reportage
44
8.8%
B
Press editorials
27
5.4%
C
Press reviews
17
3.4%
D
Religion
17
3.4%
E
Skills, trades and hobbies
38
7.6%
F
Popular lore
44
8.8%
G
Biographies and essays
77
15.4%
H
Miscellaneous (reports, official documents)
30
6%
J
Science (academic prose)
80
16%
K
General fiction
29
5.8%
L
Adventure fiction
24
4.8%
M
Science fiction
6
1.2%
N
Adventure fiction
29
5.8%
P
Romantic fiction
29
5.8%
R
Humour
9
1.8%
500
100%
Total
16/07/2005
CL 2005, Birmingham
Passives in English (1)

Be vs. get-passives
–
–
Be-passives occur in both dynamic and stative situations
Get-passives occur only in dynamic situations

–
Only be-passives are appropriate in infinitival complements

–
they liked to be/*get seen to go to church
Be-passives are predominantly more frequent than get-passives

–
Go and get/*be changed!
955 vs. 31 instances per 100K words in FLOB/BNCdemo
Be-passives are more frequent in writing while get-passives are
more frequent in spoken data

Normalised frequencies (per 100K words)
–
–
5
Be-passives: 854 in FLOB and 101 in BNCdemo
Get-passives: 5 in FLOB and 26 in BNCdemo
16/07/2005
CL 2005, Birmingham
Passives in English (2)

120
Long vs. short passives (1)
–
100
Pecentage
80
60
–
40
Agent type
20
agentless
0
agent
BNCdemo
FLOB
Corpus
6
16/07/2005
For both be and getpassives, short forms are
much more frequent than
long forms in written as well
as spoken data
Short passives are
significantly more common
in spoken than written
English

LL=209.225 for 1 df,
p<0.001
CL 2005, Birmingham
Passives in English (3)

120
Long vs. short passives (2)
–
100

Percentage
80
–
60
40
Agent type
agentless
0
agent
be-passive
get-passive
LL=76.015 for 1 df, p<0.001
The agents in get-passives are
typically impersonal (e.g. got
caught by the police) or even
inanimate (e.g. got knocked
down by a car)

20
When personal agents appear,
they are typically
informationally dense and thus
semantically indispensable
–
Passive type
7
Get-passives are more likely
than be-passives to occur
without an agent
16/07/2005
e.g. The bleeding fat girl, he
got asked out by her.
CL 2005, Birmingham
Passives in English (4)

120
Adverbials in be and get-passives
–
100

–
80
Percentage
Adverbials are more frequent in be
than get-passives
Types of adverbials are less varied
in get than be-passives

60
–
40
Adverbial type
No adverbial
0
Adverbial
be-passive
get-passive
Passive type
8
–
17.3% vs. 19.5% in FLOB and
BNCdemo
Proportion of get-passives with an
adverbial is greater in writing than in
speech

16/07/2005
Typically they ‘have an intensifying
or focusing role’ in get-passives
(Carter and McCarthy 1999: 53)
Proportions of be-passives with an
adverbial are similar in writing and
speech

20
17.7% for be-passives and 7% for
get-passives
15.2% vs. 6.6% in FLOB and
BNCdemo
CL 2005, Birmingham
Passives in English (5)

Semantic and pragmatic properties (1)
–
Get-passives are frequently used to indicate speaker
attitude towards the events described (typically a negative
evaluation) while be-passives do not appear to be used in
this way
Passive type Negative
Positive
Neutral
Be-passive
4.7%
80.3%
3.4%
58.9%
15%
Get-passive 37.7%
9
16/07/2005
CL 2005, Birmingham
Passives in English (7)

Semantic and pragmatic properties (2)
–
Collocations (L0-R1, z score>3.0, frequency>3) of
get-passives are more likely to show an inflictive
meaning than be-passives


Get-passive: 46.5% (BNCdemo) and (married in FLOB);
be-passive: 27% (BNCdemo) and 8% (FLOB)
However, get-passives are NOT necessarily more
frequently negative in spoken English
–
10
Negative instances: FLOB: 45.8%; BNCdemo: 37.3%
 Exceptionally high co-occurrence frequency of a few
neutral verbs, e.g. married , paid , dressed , changed
16/07/2005
CL 2005, Birmingham
Passives in English (8)
–
Semantic and pragmatic properties (3)

Collocations reveal that get-passives are more informal in style
than be-passives
–
The get-passive is more restricted in collocations and is likely to
co-occur with verbs referring to daily activities and informal
expressions (based on BNCdemo)
 GET - dressed, changed, get weighed, fed (i.e. eat), washed,
cleaned
 GET - pricked, hooked, mixed (up), carried (away), muddled
(up), sacked, get kicked (out), stuffed, thrown (out),
chucked, pissed, nicked
– These verbs are rarely found among the top 100 collocations for
the be-passive in BNCdemo
11
16/07/2005
CL 2005, Birmingham
Passives in English (9)

Variations across text categories
–
Passive type
Be-passives are over 8 times as
frequent in FLOB as in BNCdemo

be-passiv e
get-passiv e
20.00
Percent


10.00
–
Get passives typically occur in
colloquial and informal genres

0.00
A
C
B
E
D
G
F
J
H
L
K
N
M
R
P
S
Genre
12
16/07/2005
Text categories A-J typically show
higher proportions of be- passives
than K-R
In written genres, official documents
(H) and academic prose (J) show
exceptionally high proportions of bepassives
Biber’s (1988) MDA: be-passives
(long and short) positively weighted
on D5 (abstract vs. non-abstract
information)

Get-passives are over 5 times as
frequent in BNCdemo as in FLOB
In writing, skills/trades/hobbies (E)
and humour (R) show exceptionally
high proportions of get-passives
CL 2005, Birmingham
Passives in English (10)

Syntactic functions
–
Finite vs. non-finite positions


–
English passives are by far the most frequent in the
predicate position

–

13
97% for be-passives and 96% for get-passives
Non-finite forms

–
Finite: predicate
Non-finite: adjectival, adverbial, complement, object, subject
relatively common in object and complement positions
Rare in the subject position
The distribution of get-passives across syntactic functions is
more balanced than that of be-passives
16/07/2005
CL 2005, Birmingham
Passives in Chinese (1)

Syntactic vs. lexical passives
–
Syntactic passives



–
Lexical passives: ai, shou, zao

14
bei: most frequent and ‘universal’ passive marker
gei, jiao, rang: not fully grammaticalised, colloquial and
dialectal
Wei(-agent-)suo: archaic and typically found in formal
written genres
Lexical meanings are inherently passive
16/07/2005
CL 2005, Birmingham
Passives in Chinese (2)

Long vs. short passives
–
–
–
–
–
–
15
Bei and gei are found in both long (40%, 43%) and short (60%,
57%) passives
Wei(-agent-)suo, jiao and rang only occur in long passives
Shou and zao are more frequent in short (68%, 63%) than long
(32%, 37%) passives
Ai typically occurs in short passives (97%)
In lexical passives, the agent NPs can be systematically
interpreted as attributive modifiers of nominalised verbs, but they
cannot in syntactic passives
Long passives tend to be used in speech and colloquial genres
while short passives are found in typical written genres such as
academic prose (J), official documents (H) and biographies (G)
16/07/2005
CL 2005, Birmingham
Passives in Chinese (3)

Syntactic functions
–
Most frequent in the predicate position

–
Non-predicate positions: attributive, adverbial,
nominal, object, subject



16
76% for syntactic passives (bei 74%); 75% for lexical
passives
The attributive use is the second most important
syntactic function (14%)
Rare in the subject position
Not found as a complement
16/07/2005
CL 2005, Birmingham
Passives in Chinese (4)

Interaction between passives and aspect
–
Chinese passives are closely allied with aspect

–
–
–
–
17
syntactic passives convey an aspectual meaning of result
Bare passives account for the largest proportions for
syntactic (40%) and lexical (78%) passives
Perfective -le is frequent in both syntactic (17%) and lexical
(11%) passives
RVCs and resultative de-structure are more common in
syntactic passives while bare forms are more frequent in
lexical passives
Bare verbs are uncommon in syntactic passives, especially
when the passive constructions function as predicates
16/07/2005
CL 2005, Birmingham
Passives in Chinese (5)

Semantic properties
–
Chinese passives are “usually of unfavourable meanings” (Chao
1968: 703)


–
Proportions of negative semantic prosodies


–
Syntactic: gei (68%), rang (67%), bei (52%), jiao (50%), wei…suo
(19%)
Lexical: ai (100%), zao (100%), shou (65%)
Collocations of bei-passives

18
Prototypical passive marker bei derived from its main verb usage,
meaning ‘suffer’ (Wang 1957)
However, under the influence of Western languages, passives are no
longer restricted to verbs with an inflictive meaning in Chinese
51% negative, 39% neutral, 10% positive
16/07/2005
CL 2005, Birmingham
Passives in Chinese (6)

Variations across text categories
–
–
Passives are 11 times as frequent in writing than in speech
In writing, passives are most frequent in religious texts (D) and
mystery/detective stories (L), but least common in news editorials
(C) and official documents (H)

Unlike English, Chinese passives are rare in official documents (H)
and academic prose (J)
–
–
Bei-passives


19
Be-passives function to mark objectivity and a formal style but Chinese
passives do not have this function
The contrast in proportions between long and short forms is typically
less marked in 5 types of fiction, humour and speech
More frequently negative in news editorials (C), mystery/detective
stories (L) and adventure stories (N); predominantly negative in
speech; but rarely negative in official documents (H) and academic
prose (J)
16/07/2005
CL 2005, Birmingham
Contrast (1): Frequencies

Passives are nearly 10 times as frequent in English
as in Chinese
–
–
–
20
Be-passives can be used for both stative and dynamic
situations whereas Chinese passives can only occur in
dynamic events
Chinese passives typically have a negative semantic
prosody while English passives (especially be-passives) do
not
English has a tendency to overuse passives, especially in
formal writing (Quirk 1968; Baker 1985) whereas Chinese
tends to avoid syntactic passives wherever possible ()
16/07/2005
CL 2005, Birmingham
Contrast (2): Long vs. short forms


The agent NP in the long passive follows the
passivised verb in English but precedes it in Chinese
Short passives are predominant in English while long
passives are much more common in Chinese
–
–
Passives are used in English to avoid mentioning the agent
The agent must normally be spelt out in Chinese passives


When it is difficult to spell out the agent
–
–
21
But this constraint has become more relaxed nowadays
Passives are used in English
A vague expression such as ren ‘someone’ and renmen
‘people’ is often specified instead of using passives in
Chinese
16/07/2005
CL 2005, Birmingham
Contrast (3): Semantic properties

Chinese passives are more frequently used with an
inflictive meaning than English passives
–
–
–
Chinese passives were used at early stages primarily for
unpleasant or undesirable events; but the semantic
constraint on the use of passives has become more relaxed,
especially in writing
In this respect, the get-passive is closer to Chinese
passives than the unmarked be-passive, which is more
stylistically oriented
Proportions of meaning categories


22
English: neutral > negative > positive
Chinese: negative > neutral > positive
16/07/2005
CL 2005, Birmingham
Contrast (4): Syntactic functions






23
As a verb construction, the passive is most frequently used in
the predicate position in both English and Chinese
The proportion of passives used as predicates in English (over
95%) is much higher than that in Chinese (76% on average)
Passives are more frequent in the object than subject position
in both languages
Passives often function as attributive modifiers in Chinese but
as complements in English
Passives in Chinese (bei-passives in particular) are more
balanced across syntactic functions than English passives
Chinese passives in the predicate position typically interact with
aspect but in English the interaction between passives and
aspect is not obvious
16/07/2005
CL 2005, Birmingham
Contrast (5): Genre variations

Be-passives occur more frequently in informative than imaginative text
categories while get-passives are most commonly found in colloquial genres
and informal written genres
–


In Chinese, wei…suo typically occurs in formal written genres and jiao, rang
and gei in colloquial genres
Mystery/detective stories (L) and religious writing (D) show exceptionally high
proportions of passives in Chinese
–
–

Mystery/detective stories are often concerned with victims who suffer from various
kinds of mishaps or what criminals do to them
In religions, human beings are passive animals whose fate is controlled by some kind
of supernatural force
The difference in the overall distribution of passives is closely associated with
the different functions of passives in the two languages
–
–
24
Official documents (H) and academic prose (J) show very high proportions of passives
in English, but have the lowest proportions of passives in Chinese
(be-passives) marking an impersonal, objective and formal style in English
an ‘inflictive voice’ in Chinese
16/07/2005
CL 2005, Birmingham
Conclusions



25
Passive constructions express a basic
passive meaning in English and Chinese, but
they also show a range of differences
These differences are associated with their
different functions in the two languages
Comparable monolingual corpora provide a
useful tool for contrastive linguistics
16/07/2005
CL 2005, Birmingham