A ctive and passive voice 159

Active and passive voice

3.63 T h e term v o i c e is used to d e sc rib e the last m ajo r verb category to b e

co n sid ered in th is c h a p te r : th a t w h ich d istin g u ish es an active v erb p h ra se
(eg: ate) fro m a p a ssiv e o n e (eg: was eaten). W e h av e left th is to p ic to the e n d
o f th e c h a p te r fo r tw o r e a s o n s : first, th e passive c o n stru ctio n , w h en it occurs,
a p p ea rs in final p o s itio n in th e v erb ph rase ( c f T y p e D , 3.55), an d second,
voice is a c ateg o ry w h ich , as w e saw in 2.21, co n cern s n o t only v erb p h rases,
b u t o th e r c o n stitu e n ts in th e clau se, an d th ere fo re p o in ts to co n n ectio n s
b etw een th is c h a p te r a n d sectio n s in later c h a p te rs (esp 16.26#", 18.32) w h ere
th e c o n tra st b e tw e e n a c tiv e a n d passive m u st b e reconsidered in a larg e r
context.

V oice defined
3.64 V oice is a g ra m m a tic a l c ateg o ry w h ich m ak es it possible to view th e a ctio n
o f a se n ten ce in e ith e r o f tw o w ays, w ithout c h an g e in th e facts re p o rte d :

T h e b u tle r m urdered th e d etectiv e, [F c t i v e ] [ 1]

~ T h e d e te ctiv e was m urdered by th e b u tler, [ F
i v e ] [2]

A s w e see fro m [1] a n d [2], th e a ctiv e-p a ssiv e relatio n involves tw o

g ram m atical lev els: th e v e rb p h ra se, a n d th e clause. In th e fo rm er, a p assiv e
verb p h ra se (ie o n e c o n ta in in g a c o n stru ctio n o f T y p e D ) co n trasts w ith a n
activ e v erb p h ra se , w h ic h is sim p ly defined as o n e w h ich does n o t co n ta in
th a t co n stru ctio n . F o r e x a m p le :

Table 3.64

ACTIVE PASSIVE

present: kisses ~ is kissed

past: kissed ~ was kissed
modal: may kiss ~ may be kissed
perfective: has kissed ~ has been kissed
progressive: is kissing ~ is being kissed
modal + perfective: may have kissed ~ may have been kissed
modal -f- progressive: may be kissing may be being kissed
perfective + progressive: has been kissing ~ has been being kissed
modal -f perfective +
progressive: may have been kissing ~ may have been being kissed

I n th e v erb p h ra se, th e d ifferen ce b etw een th e tw o voice categ o ries is th a t

th e p assiv e a d d s a fo rm o f th e a u x iliary |  follow ed by th e p a st p a rtic ip le
(-ed p a rticip le) o f th e m a in verb .

T he active-passive correspondence
3.65 I n ad d itio n , a t th e clau se level, ch an g in g from th e a ctiv e to th e p assiv e
involves re a rra n g e m e n t o f tw o clau se elem en ts, a n d o n e ad d itio n (c /2 .2 1 ),
(a) T h e a ctiv e su b je ct b e co m es th e p assiv e F g e n t ; (b) th e a ctiv e o b ject
b ecom es th e p assiv e s u b je c t; a n d (c) th e p re p o sitio n by is in tro d u ced b efo re
th e a g en t. T h e p re p o sitio n a l p h ra se ( F g e n t | t - p h r F s e ) o f p assiv e sen ten ces
 160 V erbs and auxiliaries

is g e n era lly an o p tio n a l elem en t (but c/3 .7 1 ). T h e a ctiv e-p a ssiv e co rresp o n d ­
e n c e fo r a m o n o tra n sitiv e v erb (c /2 .1 6 ) w ith a n o bject {typically a noun
p h r a s e ) c an be seen d iag ram m atically as in Fig. 3.65:

S V O

s v pass A
Fig 3.65 The active-passive correspondence

A lte rn a tiv e ly , th e a c tiv e -p a ssiv e c o rresp o n d en ce c an be expressed by th is

f o rm u la :

n o u n p h rase! + a c tiv e v erb p h rase + n o u n p h ra se 2

~ noun p h ra s e 2 + passive v erb p h ra se + (by n o u n p h ra se ,)

T h e te rm s ‘su b ject’ a n d ‘o b ject’ in F ig 3.65 a re u sed h ere, as in 2.13 a n d 10.7,

to r e fe r to elem ents w ith a p a rticu la r stru ctu ral fu n ctio n in th e c lau se: eg th e
su b je c t h as co n co rd w ith th e verb p h ra se, a n d c o n tain s a su b jectiv e ra th e r
th a n o b jectiv e case p ro n o u n . T h u s c e rta in ch an g es a p a rt from those
re p re s e n te d in [1] a n d [2] m ay have to tak e p lace in ch an g in g a sen ten ce from
a c tiv e to passive ( c /3 .7 0 ) :

T h e m en respect her. [3]

~ S h e is-re sp e c te d by th e m en. [4]
i

B u t a lth o u g h the co rre sp o n d in g a ctiv e a n d p assiv e sen ten ces a p p e a r to be

r a d ic a lly different, th e re la tio n s o f m ean in g b etw een th e ir elem en ts rem ain
th e s a m e : for ex am p le, [1] h as th e sam e tru th value as [2], a n d [3] h as the
sa m e tr u th value as [4], In John helped M a ry a n d M a ry was helped by John,
John is in b oth c ases th e ‘p e rfo rm e r o f th e a c tio n ’, ev en th o u g h structurally,
John h a s a very Idifferent p o sitio n a n d fu n c tio n in each .

N o te [a] C o rresp o n d in g ac tiv e a n d p assiv e sentences do n o t, h ow ever, alw ays h av e th e sam e tru th
v alu e (c /3 .7 2 ).
[b] W e d istin g u ish term in o lo g ically th e agent, as defined a b o v e (an d in g ram m atical trad itio n ),
fro m th e ag entive, w h ich is a sem a n tic role ( c / 10.19) o fte n assum ed by th e su b ject o f a n activ e
se n te n c e o r by th e ag e n t o f a p assiv e sentence (c/9.50).

T h e passive auxiliaries: be and g e t

3.66 T h e p a ssiv e auxiliary is n o rm ally be. Its only serious c o n te n d er is get, w h ich
h o w e v er is not, b y m o st sy n tactic c riteria , a n au x iliary a t all ( c f 3 . 2 2 / /
 A ctive and passive voice 161

M o reo v er, g e t te n d s to be lim ite d to co n stru ctio n s w ith o u t a n ex p ressed

a n im a te a g en t:

T h e c a t g ot run over (by a bus). Ja m es got beaten last n ig h t.

G et w ith a n a n im a te ag en t is n o t, how ever, u n k n o w n :

J a m e s g ot caught (by th e police).

T h e g et-p assiv e is av o id ed in fo rm al style, an d ev en in in fo rm al E n g lish i t is

fa r less fre q u e n t th a n th e Ae-passi ve. F u rth e r exam ples a r e :

T h e h ouse is g ettin g rebuilt. [ 1]

S u ch c riticism s w ill g et trea ted w ith th e co n te m p t th ey deserv e. [2]
T h is sto ry e v en tu ally g ot translated in to E nglish. [3]

G et is m u ch m o re co m m o n as a ‘resu ltin g co p u la’ ( c f 1 6 .2 1 /0 in se n ten c es

lik e M y m other is g ettin g old, a n d it m ay b e b est analysed as su ch in se n ten c es
w h ic h look superficially like p assiv es, b u t w h ich could n o t b e e x p a n d e d b y
a n a g e n t (c /3 .7 7 ):
W e a re g ettin g bogged dow n in all sorts o f problem s. ( in f o r m a l) [4]
I h av e to g e t dressed b efo re e ig h t o ’clock. [ = ‘d ress’] [5]
I d o n ’t w a n t to g e t m ix ed up w ith th e police ag ain , ( in f o r m a l) [6]
Y o u r a rg u m e n t g e ts a b it confused here. [7]

S im ilarly g e t bored, g e t tired, g e t (very) excited, get lost, etc. G et is a d y n a m ic

co n clu siv e v e rb ( c f 4 .3 3 /0 . a n d th e p articip les in these sen ten ces a re s ta tiv e
(eg: dressed’ m ea n s ‘in a s ta te o f w earin g clothes’, as in Jane is already
dressed). H e n ce th e m ea n in g o f su ch sentences is p re d ic ta b le if th ey a r e
re g ard e d as S V C sen ten ces w ith g e t as m ain verb a n d th e p a rtic ip le (w ith its
m o d ifiers if an y ) as c o m p lem en t, as suggested by th is e x c h a n g e :

A : I [S] ’m [V] co m p letely co n fu sed [C],

B : Y es, I [S] ’m g e ttin g [V] co n fu sed [C], as w ell [A],

I t is im p o rta n t, th ere fo re, to d ra w a d istin c tio n b etw een th e p assiv e se n ten c es

[ 1 - 3 ] a n d th e c o p u la r sen ten ces [ 4 -7 ], w h ich w e m ay call p s e u d o - p s s i v e ( c f
F

3.77). A t th e sam e tim e , th ere is a n affinity b etw een these sen ten ce types, a n d
th is is e v id e n t w h en w e c o n sid er th e m ean in g o f th e get-passive, w h ich , lik e
th e co p u la r g et, p u ts th e e m p h asis o n th e su b ject ra th e r th a n th e a g en t, a n d
o n w h a t h a p p e n s to th e su b ject as a resu lt o f th e event. C o m p a re H e g o t
taught a lesson [‘it serv ed h im rig h t’] w ith :

He | -J taught a lesson o n th e su b ju n ctiv e (by o u r new teach er).

I t is p re su m ab ly becau se o f th e em p h a sis w h ich g e t p laces o n th e su b je ct

re fe re n t’s c o n d itio n (usually a n u n fa v o u ra b le c o n d itio n ) th a t th e a g e n t is less
u su al w ith a g et-p assiv e. T h is sam e e m p h a sis m ay acco u n t in p a r t fo r th e fa c t
th a t th e g et-p assiv e o ften reflects a n u n fav o u rab le a ttitu d e to w a rd s th e
a ctio n :

H o w d id th a t w indow g et o p e n ed l

ty p ically im p lies ‘I t should h a v e b e e n left s h u t!’

 162 V erbs and auxiliaries

N o te [a] T h e g e/-p assiv e p ro v id e s a c o n v e n ien t w ay o f av o id in g th e p assiv e w ith be in c a ses w h ere

th e r e is a p o te n tia l c o n fu sio n b etw e en th e n o rm a l p assiv e in te rp re ta tio n a n d th a t o f th e ‘s ta ta l
p a s s iv e ’ 0 /3 .7 7 ) . T h u s th e am b ig u ity b etw e en stativ e a n d d y n am ic m e an in g in The chair was
b ro k e n is elim in a te d in T he chair g o t broken .
[b] T h e id io m atic e x p ressio n g e t started is ano m alo u s, in th a t sta rt c a n n o t be tre a te d a s a
tr a n s i tiv e v erb in th is c o n te x t:
H o w soon c a n w e g e t s ta r te d o n th e sw im m ing pool ?
T h is id io m a p p e a rs to b e closely related to th e ca u sativ e use o f g e t , eg i n :

H o w soon c a n w e g e t s ta rte d o n th e po o l? ( c f 16.54)

C o m p a r e a sim ilar p seu d o -p assiv e co n stru c tio n w ith be (c/3 .7 7 N o te [a]).

[c] A p a r t from g e t, v erb s in pseu d o -p assiv e sen ten ces in clu d e become, grow, a n d seem ( c f 16.21,
16.23).

V o ic e constraints
3.67 A lth o u g h it is a g en eral ru le th a t tra n s itiv e v e rb sen ten ces c a n be e ith e r
a c tiv e o r p assiv e , th e re are a n u m b er o f ex cep tio n s w h ere th e a ctiv e
(tra n s itiv e ) a n d p a ssiv e sen ten ces a re n o t in sy stem atic c o rresp o n d en ce. W e
w ill d is tin g u ish five k in d s o f 'v o ice c o n s tra in t’ asso ciated resp ectiv ely w ith
th e v e rb (3.68 f ) , th e o b jec t (3.70), th e a g e n t (3.71), m ea n in g (3.72), a n d
fre q u e n c y o f u se (3.73).

V e r b c o n s t r a in ts
3.68 ( a ) ACTIVE ONLY
T h e r e a re g re a te r re stric tio n s o n v erb s o ccu rrin g in th e p assiv e th a n o n v erb s
o c c u rrin g in th e activ e. In a d d itio n to c o p u la r an d in tra n sitiv e v erb s, w h ic h
h a v in g n o o b je c t c a n n o t tak e th e p assiv e, som e tra n s itiv e v erb s, called
‘m id d le ’ v erb s, d o n o t o c c u r a t least in som e senses in th e p assiv e (c/T 0 .1 4 );
fo r e x am p le:

T h e y have a n ice house. T h e d ress becomes her.

H e lacks confidence. Jo h n resembles his fa th er.
T h e a u d ito riu m holds 5000 p e o p le .\ W ill th is suit you ?

A ll th e s e b elo n g to th e sta tiv e class o f v erb s o f ‘b e in g ’ a n d ‘h a v in g ’ ( c / 4.31).

B u t o th e r s ta tiv e verbs, su ch as th o se o f v o litio n o r a ttitu d e , can e asily o ccu r
in th e p assiv e. C o n tra s t:

T h e c o at d o es n o t fit you. ~ *Y ou are n o t fitted by th e coat.

T h e policfe w a n t h im . ~ H e is w a n ted b y th e police.

(b) PASSIVE ONLY

C o n v ersely , w ith som e v erb s a n d v erb c o n stru ctio n s o nly th e p a ssiv e is
p o ssib le (c /1 6 .5 0 ):

O th e r ex am p les are be born (w ith a n irreg u lar p a s t p a rticip le), a n d be drow ned
(in c ases w h ere n o a g en t is im p lied ):

H e w as b o r n in T u b in g e n . ~ ?H is m o th e r b o re h im in T u b in g e n .
 A ctive and passive voice 163

T h e w an ted m a n fell in to th e w a ter an d w as d row ned. ~ . and

som eone d ro w n e d him .

P r e p o s itio n a l v e r b s
3 .69 In E nglish, p re p o sitio n a l v erb s ( e /1 6 .5 # ) can o ften o c cu r in th e passiv e, b u t
n o t so freely as in th e a ctiv e. T h ese p rep o sitio n al v erb s a re v erb al id io m s
co n sistin g o f a lex ical v e rb follow ed by a p rep o sitio n , such as look at.
C o m p are th e follow ing se n ten c es, in w hich [ 1a] an d [2a] co n ta in p re p o sitio n al
verbs, w hereas [lb ] a n d [2b] c o n ta in th e sam e w ords in n o n id io m a tic u s e :

T h e en g in eers went v e ry carefully into jjjj® tu n n d ” 1 [lb ]

~ { ?*T he tu n n e | m | was v er^ care ^ y S one 'nt0 ^ engineers.

. . . . . . . f th e expected result. [2a]

T h ey ev en tu ally arrived a t | thg spjgn d jd sta d iu m . [2b]

f T h e e x p e c te d resu lt 1 . ,, . . .
~ i 0 *1 . 1, i j . j - j- ] w as eventually arrived at.
[ ?*T he s p le n d id sta d iu m J

In th ese sen ten ces it is c le a r th a t th e difference in a ccep tab ility c a n be s ta te d

in term s o f c o n c re te /a b s tra c t p assiv e subjects. I t is o nly in th e a b strac t,
figurative use th a t go into, arrive at, look into, a n d m an y o th e r expressions
acc ep t th e passive.
W ith som e in g en u ity , h o w ev er, one m ay c o n stru ct co n tex ts w here su ch
v erb al expressions w ill o c c u r in th e passive ev en w h ere they a re n ot used
ab stractly an d id io m a tic ally , especially in a c o o rd in ate c o n stru c tio n :

T h is p riv ate d ra w e r o f m in e h as been gone into a n d ru m m ag ed so m an y

tim es th a t it is to ta lly d isarran g ed .
T h u s th e d istin c tio n b e tw e e n p rep o sitio n al verbs a n d n o n id io m atic c o m b i­
n a tio n s o f verb a n d p re p o sitio n does n o t en tirely d e te rm in e th e p o ssib ility o f
a passive. R a th e r, th e re is a scale o f ‘cohesion’ b etw een v erb a n d p rep o sitio n ,
w h ich will b e a n aly sed la te r, in 1 6 .1 2 # w h en w e com e to e x am in e
p rep o sitio n al v erb s in m o re d etail.

O b j e c t c o n s tr a in ts
3.70 T ran sitiv e v erb s c a n b e follow ed eith e r by p h ra sal o r b y clau sal o b jects ( c f
15.3#, 16.30#). W ith cla u se s as objects, how ever, th e p assiv e tran sfo rm a tio n
is to a g reater o r lesser d e g re e re stric te d in u s e :

(A ) N o u n p h ra se a s o b je c t:
Jo h n lo v ed M a ry . ~ M a ry w as loved (by Jo h n ).
(B) C lause as o b jec t:
(i) F in ite c lau se:
Jo h n th o u g h t (th a t) she was attractive.
~ l* T h a t she w as attractive w as th o u g h t (by Jo h n ).
(ii) N o n fin ite c lau se:
(a) infinitive:
Jo h n h o p ed to m e et her. ~ * T o m eet her w as h o p ed (by John).
 164 V e rb s and auxiliaries

(b ) p a r tic ip le :
J o h n enjo y ed seeing her.
~ * I Seeing her w as enjoyed (by John).

T h e p a s s iv e o fte n b ecom es a ccep tab le, h ow ever, p a rticu la rly w h en th e o bject

is a fin ite c lau se, if th e clau sal o b ject is ex trap o sed a n d rep laced b y the
a n tic ip a to r y p ro n o u n it ( c /1 6 .3 4 ,16.72f 16.83,18.33):

I t w a s th o u g h t th a t she was attractive,

l i t w a s h o p e d to m eet her.

o r i f t h e su b je ct o f th e o b ject clause is m ad e th e su b je ct o f a passive

su p e ro rd in a te clau se (c/1 6 .5 0 /, 18.36), as in :

S h e w a s th o u g h t to be attractive.

T h e c o n s tru c tio n w ith a n tic ip a to r)' it n e v er occurs w ith p a rtic ip le clauses as

su b je ct, a n d is o n ly so m etim es a ccep tab le w ith infinitive c lau ses (c/1 6 .3 8 ):

I t w a s d e sire d to have the report delivered here.

(E v e n th e n , th e a cc ep ta b ility o f th is co n stru ctio n m ay be c riticize d on

g ro u n d s o f style.)
C o re fe re n c e b e tw ee n a su b ject a n d a n o u n p h rase o b ject b lo ck s th e passive
c o rre sp o n d e n c e . T h is c o n stra in t occurs w ith (a) reflexive p ro n o u n s, (b)
re c ip ro c a l p ro n o u n s, a n d (c) possessive p ro n o u n s w h en co refe ren tial to th e
s u b je c t:

(b) W e c o u ld h a rd ly see each other in th e fog.

~ *Each other could h a rd ly be seen in th e fog.

M y h an d
w as s h a k e n by th e w om an.
I* H e r head

N o te [a] T h e pa ssiv e o f th e sen ten ce w ith a recip ro cal pro n o u n ca n be m a d e ac c e p ta b le if th e pronoun

is s p lit in to tw o p a r ts as fo llo w s:
E a c h co u ld h ard ly be seen b y th e o th e r, ( c f 6.31)
[b] S in c e reflexive a n d recip ro cal p ro n o u n s are o b jective case rep lacem en ts, w e w ould in an y
case n o t e x p e c t th e m to ta k e su b ject p osition in a p assive sentence, o r fo r th a t m a tte r in any
o th e r s e n te n c e (c/3 .6 5 ).
fcj T h e p a ssiv e is n o t possible fo r m a n y idiom s in w h ich th e v erb a n d th e o b je ct form a close
u n it:
T h e sh ip set sail. ~ *S ail w as set.
W e c h a n g e d buses. ~ ‘ B uses w ere changed.

A g e n t c o n s t r a in t s
3.71 U n lik e th e a c tiv e subject, th e a g en t 6y-phrase is gen erally o p tio n a l. In fact
a p p ro x im a te ly fo u r o u t o f five E n g lish p assiv e sentences h a v e n o expressed
 A ctive and passive voice 165

ag en t. T h is o m issio n o ccu rs especially w hen th e ag en t is irre le v a n t o r

u n k n o w n , as in :

T h e P rim e M in iste r h as o ften been criticized recently.

o r w h ere th e a g e n t is left o u t as re d u n d a n t:

Ja c k fo u g h t M ich ael la s t n ig h t, a n d J a c k was beaten.

A n a g en t p h ra se b y M ichael would clearly n o t be necessary o r ev en fu lly

a cc ep ta b le in th is co n tex t.
S ince th e a g e n t p h ra se is usually left unex p ressed , the id en tity o f th e a g e n t
m ay b e irrec o v era b le, a n d it m ay be im p o ssib le to p ostulate a u n iq u e a c tiv e
clau se c o rre sp o n d in g to th e passive o n e :

O rd e r ha d been restored w ith o u t bloodshed.

f C o lo n el L av a l (?) "|
~ < T h e a d m in istra tio n (?) > had restored o rd e r w ith o u t b lo o d sh ed .
( T h e a rm y (?) J
N o te In so m e sen te n ces, th e ag e n t is n o t o p tio n al; eg:
T h e m u sic was fo llo w e d by a sh o rt in terv al. [1]
(*T h e m u sic w as follow ed. [2]

{TThhee rebels were actuated by b o th religious a n d political m otiv es.

rebels were actuated.
[3]
[4]
H o w ev er, th e n o n o cc u rre n ce o f th e agentless sen ten ces [2] a n d [4) m ay be du e n o t so m u c h to
g ra m m a tic a l o r lexical restrictio n , as to th e fact th a t w ith o u t th e agent, th e s en te n ce b ec o m es
in fo rm atio n ally v acuous. (In th is connection, [2] m ay be co n trasted w ith We're being
fo llow ed.)

M e a n in g c o n s t r a in ts
3.72 W e c a n n o t assu m e th a t m atc h in g activ e a n d p assiv e sentences alw ays h a v e
th e sam e p ro p o sitio n a l m ean in g . T h e d ifferen ce o f o rd er b ro u g h t a b o u t by
c h an g in g a n a ctiv e se n ten ce in to th e p assiv e o r vice versa m ay w ell m a k e a
d ifferen ce n o t o nly in em p h a sis ( c f 18.3jf, 18.1 I ff) , b u t also to th e sco p e o f
neg ativ es a n d q u a n tifiers ( c /2 .5 4 /) :

E v ery schoolboy know s one jo k e a t least. [1]

~ O n e jo k e a t least is known by ev ery schoolboy. [2]

T h e m o st lik ely in te rp re ta tio n o f [1] is q u ite different fro m th e m o st lik ely

in te rp re ta tio n o f [2]: w h ereas [1] fav o u rs th e read in g ‘E ach schoolboy k n o w s
a t least som e jo k e o r o th e r’, [2] favours th e read in g ‘T h ere is o n e p a rtic u la r
jo k e w h ic h is k n o w n to every schoolboy’.
M o reo v er, a s h ift o f m o d al m ean in g m ay accom pany a sh ift o f v o ice in
verb p h ra se s c o n ta in in g m o d al au x iliaries ( c f 4.52):

J o h n c a n n o t d o it.
~ I t c a n n o t b e d o n e (by John).

In th e a ctiv e, can here will norm ally b e in te rp rete d as ex p ressin g ab ility ,

w h ereas in th e p assiv e it is in terp reted as expressing possibility. E v en w h en
it m ig h t b e arg u ed th a t can re ta in s th e sa m e m ean in g o f ab ility in b o th a c tiv e
a n d p assiv e, a s h ift o f m ea n in g is d e te c ta b le :
 166 V erbs and auxiliaries

J o h n c a n ’t b e tau g h t. [‘I t ’s im possible to tea ch h im ’ O R ‘H e is u n ab le to

le a rn ’]
~ She c a n ’t te a c h Jo h n . [‘She is u n a b le to te a c h J o h n ’]

E x a m p le s w ith o th e r m o d al a u x iliaries a r e :

E v ery o n e o f th em m u st b e re p rim an d e d .

{ [‘E v ery o n e o f th em is to b lam e ’]

~ Y ou m u s t re p rim a n d every one o f th em .

[‘I t ’s y o u r d u ty to d o so’]

f W h y w o u ld n ’t M ira n d a rid e th e grey m are?

[‘W h y d id M ira n d a refu se?’]
- W hy w o u ld n ’t th e grey m are be r id d e n by M ira n d a ?
[‘W h y d id th e m a re refu se?’]

N o te [a] T h e sh ift from a c tiv e to p assiv e m ay ch an g e th e m e an in g n o t o nly o f a m odal co n stru ctio n ,

b u t a lso o f th e p e rfe c tiv e asp ect:
W in sto n C h u rc h ill h a s tw ice visited H arv ard . [3]
H a r v a r d h a s tw ic e b een v isited by W inston C h u rch ill. [4]
I t h a s b e e n claim ed th a t th e ac tiv e sentence [3] c a n only b e ap p ro p riately used in th e lifetim e o f
C h u rc h ill, since th e su b je c t o f th e sentence d eterm in es th e in te rp re ta b ility o f th e perfectiv e in
te rm s o f a perio d o f tim e le ad in g u p to th e p resen t ( c f 4.20). T h e p assiv e sen ten ce [4], acco rd in g
to th is claim , could a p p ro p ria te ly be said now , a fte r C h u rc h ill’s d e a th , sin ce H a rv a rd U niv ersity
is still in existence. H o w e v e r, sp eak ers hav e differing in tu itio n s on th is m a tte r.
[b] S o m e difference b e tw e e n th e m ean in g o f a n ac tiv e sen ten ce an d its p assiv e co u n te rp a rt h as
also b e e n n o ted in ex a m p le s such as [5] a n d [6], w h ere b o th su b ject a n d ob ject o f th e ac tiv e
se n te n c e are gen eric ( c f 5 .52ff):
B eav ers build d a m s. ^ ?D am s are bu ilt by beavers. [5]
E xcessive d r in k in g causes h ig h blood pressure. * ?H ig h blood pressu re is caused by
excessive d rin k in g . [6]
T h is differen ce, w h ic h is a difference o f preferred in te rp re ta tio n o nly, arise s fro m th e fact th a t
in su b je c t po sition, a g e n e ric p h ra se ten d s to be in te rp re te d u niversally, w hile in ob ject o r ag e n t
p o sitio n , this u n iv e rsa l m e an in g d isap p e ars. T hus th e su b ject beavers in [5] is likely to m ean
a p p ro x im ately ‘all b e a v e rs ’, w h ile dam s is likely to h a v e a sim ilarly g en eric m e an in g in th e
p assiv e co u n terp art.! \

Frequency co n strain ts
3 .7 3 T o th e stru c tu ra l a n d se m an tic re strictio n s m e n tio n e d in th e p reced in g
sectio n s, we m ay a d d ‘freq u en cy c o n stra in ts’. T h e re is a n o tab le difference in
th e freq u en cy w jth w h ich th e a ctiv e an d p a ssiv e voices a re used. T h e activ e
is gen erally by f a r th e m o re com m on, b u t th e re is c o n sid erab le v a riatio n
a m o n g in d iv id u a l te x t types. T h e passive h a s b een fo u n d to b e as m u ch as
te n tim es as fre q u e n t in o n e tex t as in a n o th e r. T h e m ajo r stylistic fa cto r
d e te rm in in g freq u e n c y seem s to be re la te d to th e d istin c tio n betw een
in fo rm ativ e a n d im a g in ativ e prose, ra th e r th a n to a difference b etw een
sp o k e n an d w r itte n E n g lish . T h e passive is gen erally m o re com m only used
in in fo rm ativ e t h a n in im ag in ativ e w riting, a n d is n o tab ly m o re freq u en t in
th e objective, im p e rso n a l style o f scientific a rticle s a n d new s rep o rtin g .
A s m ig h t be e x p ec te d , th e passive becom es very m u ch ra re r in c o m b in atio n s
w ith o th er co m p le x v erb c o n stru ctio n s (eg T y p es A D , B D , C D , A C D in
3.54). T h is r a rity is e x trem e in th e c o m b in a tio n s B C D o r A B C D (‘p erfectiv e
progressive p a ssiv e ’ a n d ‘m o d al p erfectiv e p ro g ressiv e p a ssiv e ’), p e rh ap s in
p a rt because o f a n a v o id a n ce o f th e a w k w ard n ess o f th e be being se q u e n ce :
 A ctive and passive voice 167

T h e C o n serv ativ es won th e election. {1]

~ T h e e lectio n w as won by th e C onservatives. [2]

B u t:

T h e C o n serv ativ es have not been winning seats lately. [3]

~ (?)S eats have n ot been being won by th e C o n serv ativ es lately.
<rare> [4]
A s th e p assiv e p a ra p h ra s e o f [3], [4] offers n o p a rtic u la r difficulties o f
in te rp re ta tio n . B u t it is lik ely to be replaced in a ctu al use by th e sim p le r
co n stru ctio n S ea ts have n ot been won by the Conservatives lately, fro m w h ic h it
differs little in m ea n in g ( c f 4 .3 8 ,4 .4 0 N o te [a]).

T h e passive gradient
3.74 T h e p u rely fo rm al d e fin itio n o f th e passive, viz th a t th e clause co n ta in s th e
co n stru ctio n be (or get) + -e d p a rtic ip le , is very b ro ad , a n d w ould inclu d e, fo r
ex am p le, all th e follow ing se n ten c es:

T h is v iolin was m ade b y m y fa th e r. [ 1]

T h is con clu sio n is h a rd ly ju stifie d by th e results. [2]
C o al has been replaced b y o il. [3]
T h is difficulty can be avo id ed in several w ays. [4]

W e are encouraged to g o o n w ith the project. [5]

L eo n ard was interested in linguistics. [6]
T h e b u ild in g is a lre ad y dem olished. [7]
T h e m o d em w orld is g ettin g [‘beco m in g ’] m ore highly
industrialized a n d m echanized. [8]

B u t ta k in g acc o u n t o f th e v e rb ’s fu n ctio n a n d m ean in g , w e p re fe r to c o n sid er

only th o se ab o v e th e b ro k e n lin e as passive. T hose below th e line, [5 -8 ], d o
n o t h a v e a c lear c o rre sp o n d e n c e w ith a n activ e v erb p h ra se o r a ctiv e c lau se,
a n d a re in creasin g ly re m o te fro m th e ‘id eal’ p assiv e o f [1], w h ich can b e
p lac ed in d irec t c o rre sp o n d e n c e w ith a u n iq u e a ctiv e c o u n te rp art. T h e
v a rie ty o f relatio n sh ip s d isp la y e d by [1 -8 ] m ay w ell be re g ard e d a s p o in ts o n
a g ra d ie n t o r scale ru n n in g fro m [1] to a sentence such as [9], w h ic h is c learly
to b e analysed a s h a v in g a n a d je c tiv a l co m p lem en t follow ing a co p u la r v e r b :

w as

{ got

seem ed
(v ery ) tire d . [9]

T h e p o ssib ility o f in sertin g very confirm s th e a d jectiv al sta tu s o f tired.

C e n tra l passives
3.7S E x am p les [ 1 -4 ] c a n be c alled ‘c e n tra l’ o r ‘tru e ’ passives. S en ten ces [ 1] an d [2]
h a v e a d ire c t a c tiv e -p a ssiv e re la tio n . T he difference b e tw ee n th e tw o is t h a t
th e fo rm er h as a p erso n al, a n d th e la tte r a n o n p erso n al a g e n t ( c /9 .5 0 ) :
~ M y fa th e r m ade th is v io lin . [ la ]
~ T h e results h a rd ly ju s tify th is conclusion. [2a]
 168 V erbs and auxiliaries

S e n te n c e [3] b rin g s som e u n clarity a b o u t th e n a tu re o f the a ctiv e c o u n te rp a rt.

T h e r e a re tw o p o ssib le activ e co u n te rp arts, d ep en d in g on th e in te rp re ta tio n
o f th e b y -p h ra s e :

~ O il has replaced coal. [3a']

~ (P eo p le in m an y countries) have replaced coal by oil. [3a"]

(T h e su p p lied a c tiv e subject, here a n d below , is given in p a ren th eses.)

I n th e fo rm er case, the by-phrase h a s b een in terp reted as a n a g e n t p h ra se
c o rre sp o n d in g to th e activ e subject, b u t in th e la tte r case, th e b y -p h rase has
b e e n g iv en an in stru m e n ta l in te rp re ta tio n (by = with). Sim ilarly am b ig u o u s
e x p re ss io n s a re b e confronted by I with a n d be impressed bylwith.
S e n te n c e [4] exem plifies th e m ost c o m m o n type o f passive, th a t w h ic h has
n o e x p ressed a g e n t (‘agentless p a ssiv e ’), an d so leaves th e su b je ct o f th e
a c tiv e c o u n te rp a rt u n d eterm in ed .

S e m i-p a ss iv e s
3.76 S e n te n c e s [5] a n d [6] re p re se n t a ‘m ix e d ’ o r sem i-passive class w hose m em b e rs
h a v e b o th v e rb al a n d ad jectiv al p ro p e rtie s (c f1 A 5 ff) . T hey a re v e rb -lik e in
h a v in g a ctiv e a n a lo g u e s:

W e are e n co u rag ed to go on w ith th e project. [5]

{ ~ (T he resu lts) en co u rag e us to go o n w ith th e project.

f L eo n a rd w as in terested in linguistics.
| ~ L in g u istics in terested L eo n ard .
[5a]
[6]
[6a]

O n th e o th e r h a n d , th e ir ad jectiv al p ro p e rties include th e p o ssib ility o f :

(a) c o o rd in a tin g th e p a rticip le w ith a n a d jectiv e;

(b) m o d ify in g t h e p a rticip le w ith quite, rather, more, etc;
(c) re p la cin g be by a lexical co p u la r v e rb such as fe e l o r se e m :
W e fe e l rather en co u rag ed a n d c o n te n t. . .
L e o n a rd seem ed very in te rested in an d keen on linguistics.

T o th e s e w e m ay ad d th e fa ct th a t [5] a n d [6] a re stativ e ra th e r th a n d y n am ic.

T h is in itself d o es n o t exclude a passive analysis, for th ere are s ta tiv e p assiv es
as w ell a s d y n a m ic passives, as is alread y illu strated by [2]: T his conclusion is
hardly ju stified b y the results. I t does, h o w ev er, tilt th e scales in fa v o u r o f an
a d je c tiv a l analysis, since all p a rtic ip ia l adjectives have a s ta tiv e m ean in g ,
w h e rea s c o rre sp o n d in g v erb s usually d o n ot.
In s u c h ad jectiv al uses o f th e p a st p a rtic ip le , it is rare to h a v e a by-phrase
ex p ressin g the a g en t, b u t b len d s su ch a s th e follow ing do o ccu r:

I feel ra th e r let d ow n by his indifference.

S h e seem s e x trem ely e lated by her success.

E v en -e d ad jectiv es w h ich h a v e no co rresp o n d in g active in fin itiv e o r finite

v e rb fo rm s m ay o ccasio n ally h a v e a g en t by-phrases:

W e w ere unim pressed by his a tte m p ts,

f *H is a tte m p ts unim pressed us.
~ [ H is a tte m p ts d id n o t im p ress us.
 A ctive and passive voice 169

E v id en tly th e a b ility to ta k e an a g en t iy -p h ra s e c a n n o t be reg ard ed as

d iag n o stic o f th e p a ssiv e co n stru ctio n . (If fu rth er c o n firm atio n is n eed ed , it
will be fo u n d in n o u n p h ra ses w here th e a g en t 6y-phrase occurs as
p o stm o d ifier: po em s b y W ordsworth.)
T h ere are, in fa c t, several p rep o sitio n s w h ich c a n in tro d u ce ag en t-lik e
p h ra ses; n o tab ly about, at, over, to, a n d with {c f 9.49f f , 16.69):

W e w ere all w o rried about the complication. [10]

~ T h e co m p lica tio n w o rried us all. [10a]
I w as a b it su rp rise d a t her behaviour. [11]
~ H e r b e h a v io u r su rp rised m e a b it. [11a]
Y ou w o n ’t b e b o th ere d with m e an y m ore. [12]
~ I w o n ’t b o th e r you an y m ore. [12a]
T h is e d itio n w as n o t k n o w n to earlier scholars. [13]
~ E a rlie r sc h o lars d id n o t k n o w (of) th is ed itio n . [13a]

B ut ju st as a fey-phrase (as n o ted in 3.75) m ay cooccur, in an in stru m e n tal

fu n ctio n , w ith a n a ctiv e subject, so th ese ag en t-lik e p h ra ses m ay so m etim es
cooccur w ith a n a c tiv e su b ject, a n d so be in te rp rete d am biguously w h en in
th e p assiv e:

L eo n ard w as in te re s te d in linguistics.
~ (S om eone) in te rested L eo n ard in linguistics.

T h ere is th u s n o stro n g reaso n to tre a t such p re p o sitio n al p h rases, w h e th er

in tro d u ced by by o r som e o th e r p rep o sitio n , as d iag n o stic o f th e p assiv e
voice.

N o te [a] Sem i-passive co n stru c tio n s su ch a s those o f [1 0 -1 3 ] ca n h av e a clause as co m p lem en tatio n

(c/16.71):
I w as su rp rised th a t th e fo o d was so good. [14]
In such cases, th e clau sal co m p lem en tatio n ca n also be seen as analogous to an a g e n t; cf:
~ That the fo o d was so good su rp rised m e. [ 14a]
[b] B e know n (to) differs fro m o th e r ex am p les o f th e sem i-passive in la ck in g th e c a u sativ e featu re
o f be worried (about), be surprised (at), etc, an d in not ta k in g m odifiers like quite an d rather :
I w as ra th e r su rp rise d a t h e r m ethods.
but:
* H er m e th o d s a re ra th e r kn o w n to me.
(O n th e o th e r h a n d , th e be know n (to) co n stru ctio n can be intensified by w ell ; b u t th e sequence
well known ca n b e h y p h e n a te d , a n d is p e rh a p s a com pound here.)

P s e u d o -p a s s iv e s
3.77 Finally, [7] a n d [8] h a v e n e ith e r an a ctiv e tran sfo rm n o r a p ossibility o f ag en t
a d d itio n :

T h e b u ild in g is a lre ad y dem olished. [7]

T h e m o d e rn w orld is g ettin g m o re highly in d u stria liz ed an d
m ec h an iz ed . [8]

Such ex am p les m ay b e called ‘p seudo-passives’, sin ce it is chiefly only th e ir

superficial fo rm o f v erb + -ed p a rticip le th a t reco m m en d s th em for
co n sid eratio n as p assiv es. In term s o f m ean in g , th e activ e sen ten ce
co rresp o n d in g to [7] is n o t [7a'], b u t [7a"]:
 170 V e rb s and auxiliaries

?* (S o m eo n e) a lre ad y d em olishes th e building. [7a']

(S o m eo n e) h a s a lre ad y dem olished th e b uilding. [7a"]

T h a t is, is dem olished d e n o te s a re su ltan t s ta te : it refers, like th e perfective,

to a s ta te resulting fro m th e dem olition, ra th e r th a n to th e a c t o f d em o litio n
its e lf. S u ch a c o n stru c tio n h as been term ed a ‘s ta ta l p assiv e’. In th is
c o n n e c tio n , w e n o te th a t an am b ig u ity is e v id e n t p a rticu larly in th e p a st
te n s e :
I n 1972, th e D e m o c ra ts w ere defeated.

O n th e d y n a m ic (cen tral passive) read in g , th is m e a n s ‘Som eone d e feated the

D e m o c r a ts ’; o n th e s ta ta l (copular) read in g , it m ea n s ‘T h e D e m o crats w ere
in a s ta te o f h a v in g b e en d e fea te d ’. The: first re ad in g c a n b e singled o u t by
a d d in g th e a g e n t p h r a s e :

I n 1972, th e D e m o c ra ts w ere d efeated by th e R ep u b lican s.

T h e first (passive) r e a d in g c a n also b e p ick e d o u t if w e ch an g e th e v erb to the

p ro g re s siv e a sp e ct:

I n 1972, th e D e m o c ra ts w ere b eing d efeated .

N e ith e r o f th ese tests a p p ly to th e sta ta l p assiv e c o n stru ctio n b ecause it is

e sse n tia lly c o p u lar, th e v e rb be in th is case b ein g th e co p u la ra th e r th a n th e
p a s s iv e au x iliary .
S im ila rly , th e p a rtic ip le s in [8] h ave a d je ctiv al v alu es: c o m p are industrial­
iz e d w ith industrial a n d m echanized w ith m echanical, a n d n o te th a t th ese -ed
w o rd s c a n b e used ad jectiv ally in p h ra ses lik e the industrialized world.
M o re o v e r, n o ‘p e rfo rm e r’ c a n be co n ceiv ed o f: a n industrialized world is
sim p ly a w orld th a t h a s re ac h ed a state o f in d u stria liz atio n . O n ce w e com e to
e x a m p le s like th is, w e a re firm ly in th e territo ry o f be as a co p u la rep laceab le
b y o th e r co p u la r v e rb s su ch as become, fe e l, seem , remain, e tc : She became
enraged, I f e lt cheated, etc. H ere th e -ed w o rd is a co m p le m e n t, a n d th erefo re
a d je c tiv a l.

N o te [a] T h e r e is a p seu d o -p assiv e c o n stru c tio n w ith in tra n sitiv e v e rb s o f m o tio n o r com p letio n in
w h ic h th e p artic ip le is a c tiv e ra th e r th a n passive in m e a n in g :
W h y are all th o se c a rs stopped a t th e co rn er?
By th e tim e sh e got th e re , h e r friend was gone.
1 7 /so o n be fin ish e d w ith th is jo b .
W i th m o st in tra n sitiv e v erb s, th is construction h as b een su p ersed ed by th e perfectiv e
c o n s tru c tio n , whi<?h is alm o st synonym ous. C o m p a re M a ry has com e w ith th e a rch a ic M a r y is
com e. T h e so m ew h at m e lo d ra m a tic im p erativ e B e gone m ay b e w ritte n as a single w ord Begone ,
a n d is so m etim es, lik e B ew are (c /3 .5 4 N o te), tre a te d as a n in fin itiv e: I told them to begone fr o m
m y sight.
[b] A sim ila r pseu d o -p assiv e is used w ith verbs o f p o s tu re :
G r a n d f a th e r vvcri1s a t in th e ro ck in g chair.
Y v e been stood h ere fo r a b o u t te n m inutes.
I n th is case, th e co n stru c tio n is largely synonym ous w ith , th o u g h less com m on th a n , th e
p ro g re ssiv e c o n s tru c tio n : was sitting , 've been standing , etc.
[c] T h e ‘n o tio n al p a ssiv e ’ w ith a n in tran sitiv e ac tiv e v erb , as in The clock winds up a t the back
[‘c a n b e w ound u p ’] is discu ssed u n d e r w o rd-form ation (A p p 1.54).
 A ctive and passive voice 171

S um m ary
3.78 S u m m arizin g th e p a ssiv e g ra d ie n t as exem plified in 3.74, we m ay set u p th e
follow ing c la ss e s:

I C e n tra l p a ssiv es
(a) W ith e x p ressed ag en ts: [1], [2], [3]
(b) W ith o u t e x p ressed ag en ts: [4]

II S em i-p assiv es: [5], [6]

I I I P seu d o -p assiv es
(a) W ith ‘c u rre n t’ c o p u la r v e rb s be, fe e l, look , e tc : [7]
(b) W ith ‘re su ltin g ’ co p u lar v erb s g e t , become , grow , e tc : [8]
( c f 1 6 .21-23 fo r th e term s ‘c u rre n t’ a n d ‘resu ltin g ’.)

B ib lio g r a p h ic a l n o t e

F o r g en e ra l tre a tm e n ts o f th e E n g lish v e rb , see P a lm e r (1974); A llen (1966); Joos (1 9 6 4 );

H u d d lesto n (1976a).
O n v erb in flectio n s, see H id alg o (1967); Q u irk (1970a).
O n re d u ce d a n d c o n tra c te d form s, see B lack (1977); B row n a n d M illar (1980); C h e s h ire
(1981); J o rg en sen (1979); Z w ick y (1970).
O n te n se a n d a sp e c t, see C o m rie (1976); D u sk o v a (1974); N eh ls (197 5 ,1 9 8 0 ); T e d e sc h i a n d
Z a e n e n (1981); also f u r th e r referen ces in th e B ib lio g rap h ical N o te to C h. 4.
O n th e p ro g ressiv e a s p e c t in p articu lar, see A dam czew sk i (1978); D uskova (1971b); H a llid a y
(1980); L ju n g (1980).
O n au x ilia ries in re la tio n to m a in verb s, see H u d d lesto n (1 974,1980); L a n g ack er (1978); N e y
(1981); P u llu m a n d W ilso n (1977); R eich (1968); R oss (1969); T w addell (1965).
O n specific verb s a n d sp ecial cases o f a u x ilia ry o r m a in verb classification, see C h a p in (19 7 3 );
C ollins (1978); C ro w ell (1959); H aeg em an (1980); Jaco b sso n (1974,1980); Q u irk a n d D u c k w o rth
(1961); S e p p an e n (1977); S v a rtv ik an d W rig h t (1977); T o ttie (1971, 1978).
O n m o d a l au x ilia ry v e rb s, see G re e n b a u m (1974); K alo g jera (1967); Lee (1978); M elc h ers
(1980); R ay n a u d (1977); also fu rth e r referen ces in th e B ibliographical N o te to C h. 4 . T h e
freq u en cy d a t a in 3.3 9 is d e riv e d from C o a te s (1983) a n d H ofland a n d Jo h an sso n (1982).
O n th e su b ju n ctiv e, see H aeg em an (fo rth co m in g b ); Jaco b sso n (1975); T u rn e r (1980).
O n th e im p e ra tiv e , see L e v in (1979); S tein (1976); also fu rth e r references in th e B ib lio g rap h ical
N o te to C h . 8.
O n th e p assiv e v o ic e, see B en n ett (1980); B olinger (1975b, 1978c); B uyssens (1979); C o u p e r-
K u h le n (1979); D a v id so n (1980); D u sk o v a (1971a, 1972); G ra n g e r (1981); M ih ailo v ic (1 9 6 7 );
P o ld a u f (1969); S te in (1979); S v artv ik (1966). F o r freq u en c y d a ta , see S v artv ik (1966) a n d
G ra n g e r (1981).
O th e r re le v a n t stu d ie s in c lu d e G re e n b a u m (1977a); Jaco b sso n (1965); W o n d e r(l9 7 0 ).
F o r stu d ie s re la tin g m o re p articu larly to m e an in g in th e v erb p h rase, consult th e B ib lio g rap h ical
N o te to C h . 4.