Adverbial Distinctions 241

[Word Order and Information Structure] views in light of examples that have come up during my teaching and
reading.
Adverbial Distinctions that Matter and Others that Don’t
2. Preamble: Factors Affecting Adverbial Position and Intonation
Joost Buysschaert
The introduction suggests that distinguishing between subclasses of ad-
University College Ghent
verbials will help in formulating rules that explain their positional and
prosodic potential.　While this is undoubtedly true, it is important to rea-
lise from the outset that the position and intonation of adverbials is affect-
1.  Introduction
ed by an interplay of (sometimes competing) factors, of which member-
On more than one occasion, linguists dealing with the topic of adverbs ship of a particular subclass is only one.
or adverbials have complained about the complexity of the issues in- In fact, one of the main claims of the present survey will be that a
volved.1　A major source of concern has been the heterogeneous nature of more robust account of adverbial behaviour can be given if the number of
the grammatical function class of adverbials.2　Accordingly, there have adverbial categories is drastically restricted and if the rest of the explana-
been several attempts at distinguishing subclasses of adverbials.　The tion is done in terms of a number of overriding principles.　A comprehen-
present article reviews some of the distinctions that have been made and sive analysis of the factors involved is beyond the scope of this article.　
attempts to assess them as either relevant or not relevant.　Relevance is Some of the syntactic, semantic, pragmatic and textual factors that can be
judged by the contribution that the distinction can make towards explain- held responsible for adverbial position are reviewed in Hasselgård (1996:
ing the positional potential and/or the intonation of adverbial constituents. 57-70).　A number of them will be explicitly elaborated on in the course
Adverbials were the topic of my doctoral dissertation (Buysschaert of this article, when their discussion becomes pertinent (as in sections 5.1
(1979)), part of which was published as Buysschaert (1982).　A summary and 5.10).
of my standpoint was published as Buysschaert (1987) (also reprinted as
Buysschaert (1990)).　With the exception of a couple of reviews
3.  The Traditional Semantic Subclasses
(Buysschaert (1989, 1998)) and a number of guest lectures, I discontinued
my research into this topic from the mid-eighties onwards, though it has The first set of distinctions that this paper will hold up to light are the
remained a favourite chapter in a grammar course that I have continued to semantic subclasses of Time, Place, Cause, Circumstance, Reason,
teach.　In many ways, the present article is a reassessment of my former Manner, Modality, Degree, etc.　They are the oldest and best known.　
Their advantage is that in many cases they can be easily recognised, al-
though some of the labels are ambiguous:
1
  Jackendoff (1972: 47) called the adverb “the most maligned part of speech” and (1) a. George did it for the money.　 (Reason or Purpose?)
before him Chomsky (1965: 219) had noted that adverbials are “a rich and as yet rela-
tively unexplored system.”　Similar complaints have been voiced by several authors b. Two runners were disqualified at the Special Olympics.　
since them. (Time or Place?)
2
  The word class of adverbs and the function class of adverbials have often been c. Generally people are hostile to change.　
called ‘dustbin categories.’

240

distinctions.indd 240-241 10.7.26 11:52:29 AM

 242 Joost Buysschaert Adverbial Distinctions 243

(Frequency or Viewpoint?) d.  In Chicago the other day, Big Billy got himself shot.

A serious disadvantage of the semantic classification is that it is virtually Despite its intuitive descriptive nature, then, the traditional semantic tax-
open-ended, as it is often possible to identify further subclasses on the ba- onomy of adverbials does not offer a reliable basis for explaining adverbi-
sis of meaning (popular subclasses of Time include frequency, duration, al position and intonation.
location in time, but Quirk et al. (1985: 8.72), for example, distinguish a It may be objected that a time-honoured mnemonic like the Manner-
further subcategory of time relationship as in up to that moment).3　If Place-Time-rule is a useful criterion for adverbial order in end-position,
rules have to be formulated for each (sub)category separately, the number and that it is based on the traditional semantic distinctions.　Yet for one
of rules will become unwieldy, with the risk that a number of useful gen- thing, the rule sometimes breaks down:
eralisations will be missed.
(4) a. Jane washed the car carefully in front of the garage last
It is striking, for example, that the following adverbials have a similar
Saturday. (Manner-Place-Time)
potential despite the fact that they belong to different semantic categories:4
b. Jane stayed there quietly all day.　 (Place-Manner-Time)
(2) a. {ˇFrequently,} the judges {frequently} drew the wrong c. We have a habit of starting at 9am in our department.　
con`clusions {΄frequently}.　 (Time/Frequency) (Time-Place)
b. {ˇProbably,} the judges {probably} drew the wrong
A more powerful explanation of adverbial order in end-position is relative
con`clusions {΄probably}.　 (Modality)
scope or, to use the term coined in Buysschaert (1982: 158), hierarchy of
c. {Unˇfortunately,} the judges {unfortunately} drew the wrong
modification.  The adverbial with the broadest scope tends to come right-
con`clusions {un΄fortunately}.　 (Fact-evaluating disjunct)
most:
It will be argued below that this generalisation can be captured by assign-
(5) a.  ( ((Jane washed the car carefully) in front of the garage) last
ing all three adverbials to the category of S-modifiers (section 5.2 below).
Saturday).
Conversely, some members of the same semantic class appear not to
c.  ( (We have a habit of starting at 9am) in our department).
share all of their positional potential.　Consider the different behaviour in
front-position of the following Place adverbials: This principle is more powerful because it also explains other cases that
do not involve Manner-Place-Time-sequences.5
(3) a.    *In Paris, she lives.　(where live = reside)
b.  In Paris, she bought two new hats. (6) a.  ( (Jill will join us at half past three) tomorrow).　
c.    *In the shoulder, Big Billy got himself shot. (Time1, Time2)
b.  ( (We appreciate your concern very much), of course).　
(Degree, Disjunct)
3
  The category is also mentioned in Declerck (1991: 221), who also names repetition
adjuncts alongside frequency adjuncts and makes further distinctions into definite and
indefinite instances of the various temporal adverbials.
4
  Curly brackets are used to show alternative positions. The symbol ‘`’ indicates the 5
  In the case of (4b), a further rule is needed to explain the order Place-Manner-
start of a falling tone, ‘΄’ signals the start of a low rising tone, ‘ˇ’ the start of a falling- Time.　Here, the Place adverbial is a verb complement.　As explained in section 4,
rising tone. verb complements follow the verb closely.

distinctions.indd 242-243 10.7.26 11:52:29 AM

 244 Joost Buysschaert Adverbial Distinctions 245

The conclusion remains that, even with respect to the rules on adverbial traditional adverbials as complements (individual examples of interest will
order, the semantic subclassification of adverbials does not qualify as one be found on pages 222, 224, 257 and 703) but does not present comple-
which is directly relevant to word order and intonation. ment-type adverbials as a major category.
The following are further examples of verb complements (the (a)-exam-
ples) in contrast with similar looking free modifiers (the (b)-examples):
4.  Verb Complements versus Free Modifiers
(8) a.  J oe rushed to the office.　(Marked: ?To the office Joe rushed.)
Examples (3a) and (3b) above illustrate an important difference between
b.  J oe typed two letters in the office.　(In the office, Joe wrote
adverbials that are actually objects—we may call them verb complements
two letters.)
—and adverbials that are not.　The verb to live, in the sense of ‘to reside’
(9) a.  Hilary has remained in Paris.　(Marked: ?In Paris Hilary re-
suggests a ‘location’ much in the same way as the verb to eat suggests
mained.)
‘food.’　Accordingly, in Paris in She lives in Paris is as much an object
b.  Hilary bought two hats in Paris.　(In Paris, Hilary bought
as fish is in They’re eating fish.
two hats.)
Not surprisingly, verb complements share the positional and intonation
(10) a.  Jack and Jill sat in the car.　(??In the car Jack and Jill sat.)
potential of objects, which basically means that they tend to follow the
b.  Jack and Jill discussed the situation in the car.　(In the car,
verb immediately and have less freedom to roam around.　For example, it
Jack and Jill discussed the situation.)
is equally marked to say (7a) as it is to say (7b):
(11) a.  The show lasted for two hours.　(*For two hours, the show
(7) a.    ?In ˇParis I would never want to `live lasted.)
b.    ?ˇFish I would never `eat b.  He kept nagging about the cost for two hours.　(For two
hours, he kept nagging about the cost.)
‘True’ adverbials—we may call them free modifiers, in the sense that they
(12) a.  Jake has behaved professionally.　(*Jake has professionally
are not prompted by the verbal meaning—are not similarly tied to post-
behaved.)
verbal position, as the felicity of (3b) illustrated.
b.  Jake has performed the Hamlet scene professionally.　(Jake
Grammarians have been slow to accept the relevance of this adverbial
has professionally performed the Hamlet scene).
distinction despite its importance.　Buysschaert (1982: 33-38) surveys a
number of accounts that have, some of them half-heartedly, some of them Example (12a) is undoubtedly the most controversial one in the series.　
more convincingly, recognised the radically separate status of object-like Yet, it cannot be denied that behave means ‘act in a certain manner.’　
adverbials.　The distinction has gained further recognition since then but Accordingly, the constituent that expresses this manner is bound to be a
is still insufficiently reflected in the accounts of, for example, standard verbal complement.
reference grammars.　Declerck (1991: 225) does recognise the existence The distinction between verb complements and free modifiers is a very
of obligatory space adjuncts (cf. (8) to (10) below), but sees them as just important one and its range has probably been underestimated.
one subtype of the adjuncts rather than as a major category in their own
right.　He does not mention other examples of object-like adverbials (cf.
(11) and (12) below).　A more recent major grammar book like
Huddleston and Pullum (2002) admittedly classifies a limited number of

distinctions.indd 244-245 10.7.26 11:52:30 AM

 246 Joost Buysschaert Adverbial Distinctions 247

ent tree nodes has continued since, see e.g. Frey and Pittner (1998) and
5.  S-modifiers versus V-modifiers
Frey (2003).
5.1.  Subclassifying Free Modifiers: Introductory Remarks My main criticism of the attempts to subclassify free modifiers in this
Although free modifiers—the ‘true’ adverbials—clearly behave differ- way is that confusion has often arisen between syntactic status per se and
ently from verb complements, not all free modifiers have the same posi- interfering factors like information structure and hierarchy of modification.
tional and prosodic potential.　In part, this can be explained by differences Sometimes, there has also been confusion with purely semantic and prag-
in formal make-up and hence ‘syntactic weight.’　Adverbial clauses (e.g. matic issues.　I will try to illustrate some of these aspects below.
when they left the show) are bulky and enjoy the least freedom.　They A distinction that seems to me syntactically very important is that be-
will mostly be left in end position though they may get moved to front tween S-modifiers and V-modifiers.　I consider S-modifiers to be a broad
position if they have topic-function (cf. section 10).6　Adverb phrases category of free modifiers whose defining characteristic is that they say
(time and again) often suffer the same fate though in journalese style time something about the fact, event or claim described in the clause as a
phrases regularly also occur in mid position (Attorney David Durkee this whole.　The sentences in (2) were illustrations in point; I repeat them as
morning announced during a press conference that etc.).　Mid position is (13) below with some comments:
the position of choice for unobtrusive adverbs (The Millers often com-
(13) a.  { ˇFrequently,} the judges {frequently} drew the wrong
plained about the noise).　Adverbs in -ly enjoy the greatest freedom be-
con`clusions {΄frequently}.　(The event was a frequent one.)
cause they are lightweight and carry a marker that makes their function
b.  { ˇProbably,} the judges {probably} drew the wrong
and status clear (cf. the examples in (2)).
con`clusions {΄probably}.　(The claim is a probable one.)
Syntactic weight, however, is one of the ‘overriding’ factors alluded to
c.  {Unˇfortunately,} the judges {unfortunately} drew the wrong
in section 2 and I will not go into the niceties of this factor as this may
con`clusions {un΄fortunately}.　(The fact was an unfortunate
involve reiterating rules and tendencies that are sufficiently known.
one.)

5.2.  S-modifiers The similarity in positional and intonational behaviour of these adverbials
More importantly for this article, the difference in positional behaviour is striking and is a compelling reason for uniting them in one and the
appears also to be influenced by differences in the syntactic status of the same category.　The category of S-modifiers is large.　In terms of the tra-
free modifiers.　In accounts using tree diagrams, the different status has ditional classifications, it includes, among others, adverbials of time (and
usually been described in terms of different positions in the sentence tree.　 its subclasses) and place as major event-modifiers, the modality adverbials
A number of these opinions were reviewed in Buysschaert (1979: 142- as claim-modifiers and the causality adverbials (cause, reason, purpose) as
144), but the practice of assigning different adverbial categories to differ- well as the evaluating disjuncts as fact-modifiers.
The fact that there are restrictions on the mutual order of S-modifiers
has prompted many other grammarians to speculate that different syntactic
categories are involved (with different positions in the sentence tree).　
6
  End position is the position after the object(s) or verb complement(s) or after the However, the ordering restrictions can usually be explained in terms of hi-
intransitive verb.　Front position is the position before the subject.　Mid position is the
position between the subject and the main verb.　When there are auxiliaries, adverbials
erarchy of modification (cf. section 3 above).　The time adverbial as well
may occur before, after or in between auxiliaries. as the place adverbial modify the ‘whole clause’ in (14) below and are

distinctions.indd 246-247 10.7.26 11:52:30 AM

 248 Joost Buysschaert Adverbial Distinctions 249

both S-modifiers in that sense.　Only, in the case of the time adverbial Again, an important generalisation can be captured if all the free modifiers
that ‘whole clause’ also includes the place adverbial: of this type are assigned to one and the same category.　Their behaviour
is very similar.　To begin with, V-modifiers, unlike the average S-modifier,
(14) a.  ((Jane washed the car in front of the garage) last Saturday).
feel uncomfortable in front-position, though prepositional phrases may be
There are also other apparent differences in behaviour between members found there (their position is freer) if the topic-function is very clear:
of the set of S-modifiers that can be explained in other than syntactic
(18) a.      *Beautifully Amanda sang the hymn.
terms.　Consider (15) and (16):
b.  With great zest, Amanda sang the hymn.7
(15) a. Vincent took the wrong `key ΄probably. c.      *Completely, the earthquake destroyed the building.
b.    *Vincent took the wrong key `probably. c.  With this pen-knife, they were able to slice the cake.
c. That Vincent took the wrong key is probable.
Secondly, if their syntactic weight allows it (cf. section 5.1) and if the in-
(16) a. Vincent took the wrong `key last ΄night.
formation structure of the sentence is such that they do not constitute the
b. Vincent took the wrong key last `night.
comment, V-modifiers can be found in mid-position (unlike verb comple-
c.     *That Vincent took the wrong key is/was last night.
ments; cf. (12a) versus (12b) above):
In section 10 below I will observe that claim-modifiers tend not to be
(19) a.  T
 he hymn was beautifully sung by a young girl called
comment-prone, which explains why (15b) is an unlikely message.　As to
Amanda.　(the by-adjunct is the comment)
the that-clauses, they may take the meaning of ‘the claim that’ or ‘the fact
b.  The earthquake completely destroyed the ambassador’s resi-
that’ but not ‘*the event that,’ which explains why (16c) is infelicitous.　It
dence.　(The object is rich in meaning and is therefore com-
is, in other words, semantic and pragmatic factors rather than truly syntac-
ment-prone.　The information value of the adverbial is now
tic ones that explain the different behaviour here.
lower and it feels more comfortable in mid-position than it
would in (17b).)
5.3.  V-modifiers
5.3.  A Conclusion on S- and V-modifiers
There is also a broad category of V-modifiers.　Their defining charac-
The conclusion so far is that distinguishing between the broad catego-
teristic is that they say something about the process, action or state de-
ries of S-modifiers and V-modifiers makes good sense and leads to useful
scribed in the verb.　V-modifiers often correspond to the traditional cate-
generalisations of rules for the position and intonation of a good many ad-
gories of Manner, Degree and Instrument:
verbials.
(17) a.  A manda sang the hymn beautifully.　(the singing was beauti-
ful; Manner)
b.  The earthquake destroyed the building completely.　(the de-
struction was complete; Degree) 7
  With great zest could alternatively be interpreted as an S-modifier, i.e. Amanda’s
c.  They sliced the cake with a pen-knife.　(the slicing was done move to start singing came about with great zest.　On similar cases of structural ambi-
with a pen-knife; Instrument) guity, see Buysschaert (1982: 151).　A third alternative is to classify it as a Su-modifier,
see section 7.

distinctions.indd 248-249 10.7.26 11:52:30 AM

 250 Joost Buysschaert Adverbial Distinctions 251

(21) a. She copied the text to the other document.

6. A Transitional Category between Verb Complements and
b. She copied the text.
V-modifiers?
c.                      ?She copied the text to one document but he did so to anoth-
Many linguists (including myself) like their distinctions to be clear-cut.　 er.
Yet there is always at least the theoretical possibility that some distinc- d.                   ?She copied the text to clipboard but she didn’t do so to the
tions display a cline rather than a clean break between categories.　In other document.
Buysschaert (1982: 152) I speculated that such a cline may exist between
The sentences with a question mark are at worst a bit odd but do not
verb complements and V-modifiers.　Both categories are elements that re-
seem implausible.　This would take the ‘third arguments’ nearer to the
fer to the verb.　Verb complements are suggested by the meaning of the
free modifiers.
verb (they are arguments of the relationship expressed in the verb),
There are a number of place adverbials that seem to belong to this tran-
V-modifiers are not.　There may, however, also be half-way cases.
sitional category:
The borderline cases are typically the third arguments in a relationship,
the other two being the subject and an object or verb complement.　The (22) a. He scribbled a few notes in his diary.
indirect object (whether prepositional or oblique) is an example: b. H
 e keeps the mower in the shed.
c. He lost all his jewels in the wash. (Halliday (1970: 150))
(20) a.  Sue donated a painting to Harry.
d. H
 ans saw the evening star in the West. (Bartsch (1976: 124))
It is easier to leave out the indirect object than the direct object, suggest- e. The gunmen hit Bill in the shoulder.
ing that the former is less intimately connected to the verb than the latter: f. S
 he was stabbed in the chest. (after Hasselgård (1996: 60))

(20) b.  Sue donated a painting. What these place adverbials have in common is that they do not say
c.       ?Sue donated to Harry.　(context required) where the event took place.　They are, in other words, no S-modifiers like
in Paris in the sentence She bought two new hats in Paris.　But whether
Some of the other criteria used in Buysschaert (1982: 79) to recognise
they are V-modifiers or verb complements remains unclear.
verb complements are also less convincing in the case of indirect objects
A possible argument in favour of an in-between category could be that
than in the case of direct objects:
the order of the second and the third argument seems fixed, which one
(20) d.                 *Sue donated a painting to Harry but Hilda did so a chest of could take to be a decisive word order criterion:
drawers.
(20) h.    *Sue donated to Harry a painting.
e.                    ?Sue donated a painting to Harry but Hilda did so to Stephen.
f.                      *Sue donated a painting to Harry but she didn’t do so a chest Yet this may alternatively be explained by the difference in the composi-
of drawers. tion of the constituents, where the rule could be that noun phrases precede
g.                  ?Sue donated a painting to Harry but she didn’t do so to prepositional phrases.　In fact, when neither constituent has a preposition,
Stephen. the order may be reversed for reasons of syntactic weight, witness sen-
tences like She gave him a letter.
Similar doubts arise for the third argument in (21):
The question whether a separate in-between category—which might be

distinctions.indd 250-251 10.7.26 11:52:30 AM

 252 Joost Buysschaert Adverbial Distinctions 253

called the Third Argument Adverbial—should be sanctioned remains un- same syntactic category.　The adverbial in (25a) is, in principle, a
decided.　On the one hand, one may argue that ‘donate something to V-modifier and so is that of (25c).　Yet front-position does not seem all
someone,’ ‘keep something somewhere,’ ‘hit someone somewhere’ are that impossible for attentively, which has an orientation towards the sub-
valid conceptual units—more so than ‘buy something somewhere.’　This ject, whereas it is clearly excluded for completely, which is not subject-
would be an argument for saying that the third arguments are also verb oriented.
complements but possibly somewhat less integrated than the second argu-
(25) a.  She listened to all the instructions attentively.
ment.　On the other hand, it is true that the constituents under discussion
b.           ?Attentively, she listened to all the instructions.
are much like V-modifiers; arguably, their seemingly closer relationship to
c.  The job has now been finished completely.
the verb stems from an orientation towards the second argument:
d.           *Completely, the job has now been finished.
(23) a. He keeps the mower in the shed.　(also means: the mower is
Arguably, attentively in (25b) is not used as V-modifier but rather as a
in the shed)
Subject Adjunct, a category mentioned in Quirk et al. (1972: 465ff).
b. He cleans the mower with petrol.　(does not similarly mean:
Declerck (1991: 228) also has a typical example (though he does not
the mower is with petrol)
use the term Subject Adjunct):
The contentious constituents are not unlike complements to the object in
(26) Enthusiastically, he mounted the platform and addressed the
this respect, i.e. like blue in She painted the door blue.
crowd.

He paraphrases this utterance as ‘He was enthusiastic when he mounted

7.  Subject-Oriented Adverbials, Su-modifiers
etc.’　Enthusiastically can hardly be called an S-modifier here: a para-
A second difficult case is that of subject-orientation and the question phrase like ‘The fact that he mounted etc. was enthusiastic’ would not be
whether subject-oriented adverbials constitute a class of their own.　In appropriate.　A V-modifier reading is also doubtful.　Probably, the mount-
Buysschaert (1982: 150-151) I contend that subject-orientation is a con- ing was performed in an enthusiastic manner but this is not what the
comitant feature that does not influence the position or intonation of a free speaker wants to say in the first place.　The only valid paraphrase is
modifier.　This may be true of carelessly in: Declerck’s.
It is also significant that the effect tends to disappear when enthusiasti-
(24)  Carelessly, John trod on the snail.
cally is no longer in front-position.　The versions in (27),
The utterance implies that John was careless, but this subject-orientation
(27)  He {enthusiastically} mounted the platform {enthusiastically}.
does not affect the potential position and intonation of the adverbial,
which is in the first place an S-modifier (The fact that John trod on the rather suggest the V-modifier meaning, though there is admittedly only a
snail was careless [of him]), explaining why front-, mid- and end-position thin line between the readings.
are all possible (bearing in mind that the intonation of a non-comment- The conclusion may be that, despite claims to the contrary in my earlier
prone adverbial will be needed; see section 10). accounts of adverbials, a separate category of Subject Adjuncts is in order
Yet in some cases the subject-orientation seems to make front-position after all.　Since I will not accept a category of adjuncts (cf. section 9 be-
more likely than in the case of non-subject-oriented adverbials of the low), it would be more logical to use a term that is more in line with the

distinctions.indd 252-253 10.7.26 11:52:30 AM

 254 Joost Buysschaert Adverbial Distinctions 255

terms I have proposed for other free modifiers.　I suggest using the new f. All bankers are sinners, theoretically.
term Su-modifier.
Also, I have since come across examples where the U-modifier seems un-
Su-modifiers say something about the subject but rather than giving a
problematic in mid-position after all:
more or less stable characteristic as an adjective would, they describe a
state that the subject is in while performing the activity described in the (29)  I frankly/honestly don’t know what you’re talking about.
utterance.　On the surface, Su-modifiers behave like an S-modifier with a
Possibly mid-position becomes more plausible with light subjects, when
strong topic-function, which means that they will be found in front-posi-
mid-position is not all that different in effect from front-position.　If this
tion with a falling-rising intonation.
is true, it would mean that (30) is less acceptable than (29):

(30)                ?Jim and Sam frankly/honestly don’t know what they are talking
8.  U-modifiers
about.
S-modifiers were described above as saying something about the fact,
Quirk et al. (1985: 615) and Declerck (1991: 232) call the former category
event or claim described in the clause as a whole.　This seems not to in-
(frankly, honestly) Style Disjuncts and the latter (theoretically, in theory,
clude examples like the following:
technically, etc.) Viewpoint Adverbials, which they regard as a subcatego-
(28) a.  Frankly, no-one knows what Bill has in mind. ry of Subjuncts (Quirk et al. (1985: 448), Declerck (1991: 227)).　This
b.  Theoretically, all bankers are sinners. means that they miss out on a generalisation.　Both types of adverbials
can be paraphrased using the expression ‘speaking’ ({frankly} speaking
These two adverbials comment on the clause as an utterance.　They say
{frankly}, theoretically speaking) and they share positional-intonational
that it is a frank utterance, or that it is one with a theoretical value only.　
characteristics.
Does this warrant the creation of a separate category of Utterance-
The question, however, whether they are a separate category of free
Modifiers (U-modifiers) or should we rather add these free modifiers as a
modifiers or just a somewhat special type of S-modifier remains unre-
further semantic subcategory of the S-modifiers?
solved.
In my teaching, I have so far worked with a separate category of
U-modifiers, because unlike ‘real’ S-modifiers, the examples above seem
reluctant to take mid-position: 9.  Other Categories of Adverbials
(28) c.               *No-one frankly knows what Bill has in mind. The adverbials that we have discussed so far are either verb comple-
d.              *All bankers are theoretically sinners. ments (objects, in fact) or free modifiers, though a transitional category of
Third Argument Adverbials may also be recognised.　The free modifiers
One may argue, however, that this is because frankly and theoretically are
can be further subdivided into three or four categories:
very topic-prone (see section 10 for this concept).　This makes front-posi-
tion the preferred place-to-be, with end-position—as an afterthought—the —S-modifiers
second best option: —V-modifiers
—Su-modifiers
(28) e. No-one knows what Bill has in mind, frankly.
—U-modifiers (though these might also be a subtype of S-modifiers)

distinctions.indd 254-255 10.7.26 11:52:31 AM

 256 Joost Buysschaert Adverbial Distinctions 257

All the other categories that grammarians have enumerated—and there are adverbials has been broadened into a ‘dustbin of grammar.’
many—have been brushed aside. Finally, a word may be said on the distinction adjunct-subjunct-disjunct
More categories are admittedly needed if the analysis goes beyond or originally introduced by Greenbaum (1969) and taken over in the Quirk
below sentence level.　At interclausal/intersentential level one should add grammars as well as by Declerck (1991: 214-215).　The idea behind this
the conjuncts.　They are conjunctions that masquerade as adverbials: distinction is that some adverbials, the adjuncts, are better integrated in
the clause than others, which have been called disjuncts and subjuncts.　
(31) Our staff are hopeless.　The secretary, however, does a perfect
Viewpoint adverbials and fact-evaluaters like fortunately are believed to
job.
be among the less integrated adverbials.　Detailed criticism of this distinc-
However can be found in front-, mid or end-position (always with ‘com- tion is beyond the scope of this article.　It is true that some adverbials
ma-intonation’ though), mimicking the behaviour of an S-modifier despite give additional, somewhat parenthetical, remarks rather than specific in-
its very different status as a clause- or sentence-(if not paragraph-)linker. formation; but this has to do with meaning rather than syntax and may
At phrase level, there are A-modifiers (Buysschaert (1982: 87)).　They also be influenced by context (see section 10 below).
say something about the characteristic described in an adjective or adverb Declerck takes front position with comma-intonation to be uncharacter-
(or an adjectival/adverbial phrase): istic of adjuncts, but those of his adjuncts that can occur as topics can
also take this position and intonation:
(32)  They were very old and worked very hard.
(35) ˇYesterday Mary called him a `fool and ˇnow she wants to
Their position is unproblematic: they remain close to their head.　Tra­
`marry him.
ditionally, they have also been called adverbials, though their role is rather
akin to that of an adjective within a noun phrase. Another test that is meant to potentially single out adjuncts and is illus-
A more special category that works at phrase level are the focussing trated in (36a) fails in (36b):
adverbials.　Rather than qualifying a constituent, they put it in focus:
(36) a. John arrived yesterday but Mary did not (= but Mary did not
(33) a.  Only John knows where the key lies hidden. arrive yesterday).
b.  Even Mary refused to tell us where to look. b. They take the `metro in Paris but Glaswegians still prefer the
`car to go to work (≠ but Glaswegians still prefer the car to
The constituent focussed upon always bears a nuclear accent (unless the
go to work in Paris).
focussing adverbial itself has the nucleus).　If that constituent is not the
subject, the focussing adverbial may also be found in mid-position but Degree of integration in the clause as defined in the adjunct-subjunct-dis-
nevertheless remains linked to the element that has the nucleus: junct distinction may well have to do with factors of information structure
rather than true syntactic factors.
(34) a.  John only wanted to `scare them.
b.  John only wanted to scare their `mothers.
10.  Topic-Proneness, Comment-Proneness
Focussing adverbials are an interesting category (see Nevallainen (1991)
for a historical perspective) but had better be considered as rather different Buysschaert (1982: 116-140) devotes an entire subchapter to the impor-
from other adverbials.　They are another example of how the category of tance of information structure in the discussion of adverbial position and

distinctions.indd 256-257 10.7.26 11:52:31 AM

 258 Joost Buysschaert Adverbial Distinctions 259

intonation.　A major dichotomy in information structure is the distinction tains other comment-prone elements, the chances of the adverbial to make
between the part of the utterance that is presented as the subject-matter of it as the comment will diminish accordingly:
that utterance and which is often referred to as the topic (though defini-
(39) a. A
 fter she’d beautifully sung that ancient ˇhymn, … (contrast
tions differ); and that part of the utterance which asserts something (or
with (37a): the manner in which the hymn was sung is no
asks something) about the subject-matter and which is therefore called the
longer a crucial part of the message).
comment.　Furthermore, utterances may contain parts that are neither top-
b. M
 any people take the `metro in Paris/in ΄Paris (contrast with
ic nor comment.
(37c)).
Some adverbials are likely to become comments because they hold very
specific information, others are unlikely to be comments because their in- Similarly, a number of adverbials have a meaning that makes them eligi-
formation content is low or because they constitute additional remarks ble to function as topics.　Su-modifiers, U-modifiers and Conjuncts are
rather than real assertions. among them; so are fact-evaluating S-modifiers.　They are typical sen-
V-modifiers indicating a manner in which, or an instrument with which, tence-openers and often occur in front-position with a falling-rising into-
something happens or is performed are comment-prone in this sense and nation, called the pre-clause position in Declerck (1991: 216).　Again, this
so are S-modifiers that qualify the clause as an event.　Many utterances in does not mean they constitute a separate syntactic category.
which these adverbials occur are used specifically to indicate manner or
instrument, or place, time or reason.　This explains why end-position with
11.  In Conclusion
falling intonation is very popular with such adverbials:
In this article I have tried to discuss which adverbial distinctions are
(37) a. She sang the hymn `beautifully.
useful in predicting word order and intonation, and which distinctions are
b. The janitor sliced it with a `pen-knife.
not useful.　I have argued that the traditional semantic categories (time,
c. Look, I’ve bought these in `Paris.
place, manner, etc.) are not useful, because they offer an open-ended, not
S-modifiers that evaluate the fact or assess the probability of the claim always transparent classification that misses out on a large number of gen-
expressed in the clause are additional remarks rather than central com- eralisations.　I have also discounted the adjunct/disjunct/subjunct distinc-
ments.　Their favoured place will be mid- or end-position without a nu- tion because it confuses syntactic integration with issues of information
cleus or end-position with a low rising tone (the latter is called post- structure and again misses out on generalisations.
clause position in Declerck (1991: 216)). I end up with far fewer ‘useful’ categories of adverbials than traditional
accounts have used.　To begin with, I separate verb complements from
(38) a. They’re {fortunately} not at `home {fortunately/΄fortunately}.
free modifiers.　The verb complements are actually objects, the free modi-
b. They’re {probably} not at `home {probably/΄probably}.
fiers are not similarly tied to the verb.　There may admittedly be an in-be-
It would be wrong, however, to conclude that we need two more adverbial tween category of Third Argument Adverbials.
categories, i.e. one of comment-prone adverbials and one of non-com- Within the free modifiers, those that modify the clause (S-modifiers)
ment-prone adverbials.　The likelihood that an adverbial will or will not should be distinguished from those that modify the verb (V-modifiers).　
become a comment is not a syntactic characteristic but a semantico-prag- Possibly, we should also regard those that comment on the utterance as a
matic feature which ultimately depends on context.　If the sentence con- separate category (U-modifiers), though they might also be a subtype of

distinctions.indd 258-259 10.7.26 11:52:31 AM

 260 Joost Buysschaert Adverbial Distinctions 261

S-modifiers.　In this article I also admit that some free modifiers seem to Chomsky, Noam (1965) Aspects of the Theory of Syntax, MIT Press, Cambridge,
concentrate specifically on the subject and that therefore a further category MA.
of Su-modifiers may be in order. Declerck, Renaat (1991) A Comprehensive Descriptive Grammar of English,
Kaitakusha, Tokyo.
Apart from these categories, there are also other elements that have tra-
Frey, Werner and Karin Pittner (1998) “Zur Positionierung von Adverbialen im
ditionally been called adverbials but that are of a rather different nature.　
deutschen Mittelfeld,” Linguistische Berichte 176, 489-534.
They include modifiers of adjectives and adverbs (A-modifiers); clause- Frey, Werner (2003) “Syntactic Conditions on Adjunct Classes,” Modyfing
linkers (conjuncts) and focussing adverbials. Adjuncts, ed. by Ewald Lang, Claudia Maienborn and Cathrine Fabricius-
Each of the ‘useful’ categories mentioned above has its potential and Hansen, 163-209, de Gruyter, Berlin/New York.
preferred positions and intonations.　Which of these positions/intonations Greenbaum, Sidney (1969) Studies in English Adverbial Usage, Longman,
is actually chosen depends on a number of overriding factors.　Some of London.
these have been illustrated in the article: they include syntactic weight, Halliday, Michael A. K. (1970) “Language Structure and Language Function,”
New Horizons in Linguistics, ed. by John Lyons, 140-165, Pelican,
topic-proneness and comment-proneness.　A systematic account of the in-
Harmondsworth.
terplay of these factors was beyond the scope of this article but the illus-
Hasselgård, Hilde (1996) Where and When? Positional and Functional Con­
trations that have been given will, it is hoped, be sufficiently instructive. ventions for Sequences of Time and Space Adverbials in Present-day English,
Scandinavian University Press, Oslo.
Huddleston, Rodney and Geoffrey K. Pullum et al. (2002) The Cambridge
References
Grammar of the English Language, Cambridge University Press, Cambridge/
Bartsch, Renate (1976) The Grammar of Adverbials: A Study in the Semantics of New York.
Adverbial Constructions, North-Holland, Amsterdam. Jackendoff, Ray S. (1972) Semantic Interpretation in Generative Grammar, MIT
Buysschaert, Joost (1979) An Examination of the Syntactic Status, Surface Press, Cambridge, MA.
Position and Intonation of English Adverbials, Doctoral dissertation, Ghent Nevallainen, Terttu (1991) But, Only, Just. Focusing Adverbial Change in Modern
University. English 1500-1900, Société Néophilologique, Helsinki.
Buysschaert, Joost (1982) Criteria for the Classification of English Adverbials, Quirk, Randolph, Sidney Greenbaum, Geoffrey Leech and Jan Svartvik (1972) A
Koninklijke Academie voor Wetenschappen, Letteren en Schone Kunsten van Grammar of Contemporary English, Longman, London.
België, Brussels. Quirk, Randolph, Sidney Greenbaum, Geoffrey Leech and Jan Svartvik (1985) A
Buysschaert, Joost (1987) “The Position of English Adverbials,” International Comprehensive Grammar of the English Language, Longman, London.
Review of Applied Linguistics in Language Teaching (IRAL) 25, 103-117.
Buysschaert, Joost (1989) “Review of Syntax der englischen Averbialien, by
Friedrich Ungerer, Max Niemeyer, Tübingen, 1988,” Leuvense Bijdragen 87,
230-231.
Buysschaert, Joost (1990) “The Position of English Adverbials,” Grammatical
Studies in the English Language: Studies in Descriptive Linguistics, ed. by
Dietrich Nehls, 37-51, Julius Groos, Heidelberg.
Buysschaert, Joost (1998) “Review of Where and When, by Hilde Hasselgård,
Scandinavian University Press, Oslo, 1996,” English Language and
Linguistics 2, 347-350.

distinctions.indd 260-261 10.7.26 11:52:31 AM