Maienborn Schafer 2011 Adverbsandadverbials

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/236898368
Adverbs and adverbials
Chapter · November 2011

DOI: 10.1515/9783110626391-014
CITATIONS READS
28 24,595
2 authors:
Claudia Maienborn Martin Schäfer

University of Tuebingen University of Tuebingen
68 PUBLICATIONS 1,229 CITATIONS 19 PUBLICATIONS 97 CITATIONS
SEE PROFILE SEE PROFILE
Some of the authors of this publication are also working on these related projects:
understanding compounds View project
semantics of adverbials View project
All content following this page was uploaded by Claudia Maienborn on 30 May 2014.
The user has requested enhancement of the downloaded file.

1390 XI. Semantics of adjectives and adverb(ial)s
Schwarzschild, Roger 2005. Measure phrases as modifiers of adjectives. Recherches Linguistiques

de Vincennes 35, 207–228.
Schwarzschild, Roger 2008. The semantics of comparatives and other degree constructions. Lan-
guage and Linguistics Compass 2, 308–331.
Seuren, Peter 1973. The comparative. In: F. Kiefer & N. Ruwet (eds.). Generative Grammar in
Europe. Dordrecht: Reidel, 528–564.
Sharvit, Yael & Penka Stateva 2002. Superlative expressions, context, and focus. Linguistics &
Philosophy 25, 453–504.
Stassen, Leon 1985. Comparison and Universal Grammar. Oxford: Blackwell.
Stateva, Penka 2000. In defense of the movement theory of superlatives. In: R. Daly & A. Riehl
(eds.). Proceedings of ESCOL 1999. Ithaca, NY: CLC, Publications, 215–226.
Stateva, Penka 2002. How Different are Different Degree Constructions? Ph.D. dissertation. Univer-
sity of Connecticut, Storrs, CT.
von Stechow, Arnim 1984. Comparing semantic theories of comparison. Journal of Semantics 3, 1–77.
von Stechow, Arnim 2005. Different Approaches to the Semantics of Comparison. Ms. Tübingen,
University of Tübingen.
von Stechow, Arnim 2006. Times as Degrees. Ms. Tübingen, University of Tübingen.
Szabolcsi, Anna 1986. Comparative superlatives. In: N. Fukui, T. Rapoport & E. Sagey (eds.). MIT
Working Papers in Linguistics 8. Cambridge, MA: MIT, 245–265.
Villalta, Elisabeth 2007. Comparison Constructions in Motu. Ms. Tübingen, University of Tübingen.
Sigrid Beck, Tübingen (Germany)
54. Adverbs and adverbials

1. Introduction: Towards a definition of adverbs and adverbials
2. Semantic classification of adverbials
3. Adverbials at the syntax/semantics interface
4. Theoretical approaches
5. Challenges to compositionality and ontology
6. Conclusion
7. References
Abstract
The article offers an overview of the heterogeneous set of lexical and semantic classes and
subclasses of adverbs and adverbials with their characteristic inferential and distributional
properties. Furthermore, it sketches major theoretical approaches that have been devel-
oped to account for adverbial semantics and introduces some current issues of debate
concerning the proper combination of lexical, compositional, and conceptual semantics
for adverbials.
1. Introduction: Towards a deﬁnition of adverbs and adverbials

Adverbs and adverbials are highly adaptive expressions. They arise in a variety of envi-
ronments from which they take on certain characteristic features. This makes them a very
von Heusinger, Maienborn and Portner (eds.) 2011, Semantics (HSK 33.2), de Gruyter, 1390–1420
54. Adverbs and adverbials 1391
flexible means of natural language expression. Their semantics raises some intriguing
puzzles for linguistic theory that have attracted much interest in current semantic
research as documented, e.g., by the collections in Lang, Maienborn & Fabricius-Hansen
(2003), Austin, Engelberg & Rauh (2004) or McNally & Kennedy (2008). The aim of
this article is to provide an overview outlining the major semantic issues involving
adverbs and adverbials and sketching some major theoretical approaches that have been
developed to account for adverbial semantics, as well as current issues of debate.
The article is organized as follows: The introductory section provides a characteriza-
tion of adverbs and adverbials that will serve as working base for the remainder of this
article. Section 2 lays out a classification of adverbials based on semantic criteria and
includes some remarks on the delineation of adverbials and secondary predicates. Sec-
tion 3 discusses the syntax/semantics interface addressing the relationship between the
position of adverbials and their interpretation. Section 4 presents three major formal
semantic approaches that have been developed for adverbials: the operator approach
most prominently advocated by Thomason & Stalnaker (1973), McConnell-Ginet’s
(1982) argument approach, and the nowadays widely assumed Davidsonian predicate
approach. On this basis, section 5 discusses some challenges concerning the composi-
tional semantics and the underlying ontology of adverbials that current theories address.
The article ends with a short conclusion in section 6.
Clear-cut definitions of adverbs and adverbials are difficult to formulate. Since we
define the word class adverb on the basis of the syntactic function adverbial, we will start
with the latter. Not all aspects mentioned in this definition hold for all adverbials, but it
covers most types of adverbials unambiguously treated as such in the literature.
1.1. Adverbials
The term “adverbial” refers to a specific syntactic function within a sentence and there-
fore contrasts with other syntactic functions, such as subject, object, and predicate.
Adverbials are traditionally conceived of as being those elements that serve to specify
further the circumstances of the verbal or sentential referent. They are restricted to a set
of semantically limited usages, prototypically specifying time, place, or manner, cf. the
italicized strings in (1).
(1) a. Paul laughed the whole day.

b. The children played in the kindergarten.
c. Henriette dances beautifully.
The adverbials in (1) pass standard constituency tests: They can be elicited by questions,
can be replaced by pronouns, and are movable. The type of wh-word used for elicitation
varies with the semantics of the adverbial. Temporal adverbials like the whole day in (1)
answer the question When/For how long ... ?, depending on whether they specify the time
or length of the laughing. The prepositional phrase in the kindergarten in (1) is a loca-
tive adverbial, answering the question Where ... ? Finally, beautifully in (1) is a manner
adverbial, answering the question How ... ?
As the sample sentences in (1) already show, the function of adverbials may be real-
ized by different kinds of phrasal units, here noun/determiner phrases, prepositional
phrases and adverb phrases. Other phrasal units frequently functioning as adverbials are
adjective phrases and clauses; for an overview cf. van der Auwera (1998), cf. also article
55 (Sæbø) Adverbial clauses.
The prototypical adverbial is optional and corresponds syntactically to an adjunct,
acting semantically as a modifier. Examples for subcategorized adverbials are given in (2).
(2) a. Norah treated James *(badly).

b. John behaved *(admirably).
c. New York lies *(on the Hudson river bank).
The sentences in (2) require the presence of the adverbials –note, though, that John
behaved is acceptable due to a conventionalized reading of bare behave as behave well –,
contrasting with verbs like to dress in (3), which is acceptable without an adverbial when
pragmatically licensed as in (3b); cf. Ernst (1984) and Goldberg & Ackerman (2001).
(3) a. Norah dresses #(stylishly).

b. Norah dresses, but the natives prefer to go naked.
1.2. Adverbs
The term “adverb” refers to a specific word class or lexical category and therefore con-
trasts with other word classes, such as nouns, adjectives, verbs, or prepositions.
On the one hand, both adverbs and prepositions are uninflected, with adverbs dif-
fering from prepositions in having phrasal status. Reductionist approaches have there-
fore proposed to analyze at least some adverbs as intransitive, i.e. objectless, prepositions;
e.g. Jackendoff (1972), Wunderlich (1984). While this might be a viable option for some
adverb candidates such as up, down, away, there is some consensus that such reductionist
attempts are only feasible within certain limits suggesting that a lexical category of
adverbs is needed after all; cf. the discussion in Delfitto (2000: 16ff).
On the other hand, adverbs differ from nouns, adjectives, and verbs in that they often
do not possess clear markers for category membership and can only be defined via their
syntactic function of being prototypically used as adverbials. In English, both cases exist:
There is a large class of deadjectival -ly adverbs that can be identified through their mor-
phology as adverbs. On the other hand, words like well are identified as adverbs because
they can only have an adverbial function.
For English, any further attempt to give a positive definition of the word class “adverb”
is wrought with difficulties. First of all, a subclass of English adverbs (and adverbs in
other Germanic languages) can, besides serving as standard adverbials, be used to modify
adjectives or other adverbs, cf., e.g., extremely in (4).
(4) a. He drives extremely/too/very fast.

b. an extremely/very awkward situation
This kind of usage is not restricted to traditional degree adverbs like extremely, too, and
very. A fairly large class of adverbs can be used as modifiers of adjectives, cf. (5).
(5) Joe is provocatively/disappointingly/grotesquely/remarkably stupid.

Notice that these adverbs are not parallel to the degree adverbs in (4). See Morzycki
(2008) for a detailed discussion of this point; Rawlins (2008) discusses the pre-adjectival
use of illegally.
A second difficulty concerns items like tonight, tomorrow, yesterday which are usually
considered prototypical English adverbs. These items, besides being used adverbially, can
also serve as subjects; cf. (6).
(6) a. Yesterday was a beautiful day. [Adv as subject]

b. Peter worked in his office yesterday. [Adv as adverbial]
This data is problematic insofar as we argued above that the adverbial function is the
basis for the category “adverb”. If we continue to classify items like yesterday as adverbs,
we have to accept that some adverbs can serve both as adverbials and as subjects. An
elegant solution to this problem is given in Huddleston & Pullum (2002: 564ff), who ana-
lyze yesterday and similar items as pronouns. This analysis explains their distributional
pattern, which they share with standard noun phrases, cf. (7).
(7) a. The whole year was a study in failure and disillusion. [NP as subject]
b. Peter worked in his office the whole year. [NP as adverbial]
A further characteristic of adverbs in English and German is the fact that they cannot
occur as prenominal attributive modifiers of nouns, cf. (8) for English.
(8) a. *the well runner

b. *the extremely conditions
Adverbs are often classified according to their lexical semantics, cf. e.g. (9):
(9) a. now, tomorrow, afterwards [temporal adverbs]

b. here, elsewhere, inside [locative adverbs]
c. often, seldom, frequently [frequency adverbs]
Finally, it should be noted that, cross-linguistically, the word class adverb is not frequent;
cf. Sasse (1993).
2. Semantic classiﬁcation of adverbials

Common classifications of adverbials are based on either semantic or syntactic cri-
teria, or both. Here, we will give a classification based on semantic criteria alone and
discuss the interaction of adverbial subclasses with syntax in section 3. Adverbials may
be subdivided into three major groups: predicational adverbials, participant-oriented
adverbials, and functional adverbials (these terms are adapted from Ernst 2002), which
may be roughly characterized as following: Predicational adverbials assign a (gradable)
property to the verbal or sentential referent they combine with. Participant-oriented
adverbials introduce a new entity that takes part in the eventuality described by the
verb. “Functional adverbials” is the cover term for the remaining adverbials, including
quantificational and discourse-related adverbials. Before embarking on a more detailed

discussion of these semantic subclasses, we will briefly introduce the semantic notions of
opacity and veridicality, which will turn out to be crucial devices in classifying adverbials.
Opacity
In extensional systems of logic, it is usually assumed that Leibniz’ Law holds: Two co-
referential expressions can be freely substituted for one another without changing the
truth value of the original expression. Expressions for which this law does not hold are
oblique or referentially opaque. As (10) shows, adverbials can give rise to opaque contexts:
(10) a. Necessarily, Sam Peckinpah is Sam Peckinpah.

b. Necessarily, Sam Peckinpah is the director of The Wild Bunch.
While (10a) is analytically true (in most systems of logic), (10b) is false. Adverbials can
be characterized as to whether they create opaque contexts for all positions in a sen-
tence, for just specific positions, or for no positions at all.
Veridicality
An adverbial is veridical (or factive), if a sentence containing the adverbial entails the
sentence without the adverbial. It is nonveridical, if there is no such entailment. Some
adverbials, e.g. functional adverbials like never, are antiveridical, that is, they entail that
the sentence without the adverbial is not true; cf. Giannakidou (1999) and also Bonami,
Godard & Kampers-Manhe (2004).
2.1. Predicational adverbials

Predicational adverbials can typically be characterized as supplying a gradable prop-
erty on the verbal or sentential base. (By restricting predicational adverbials to those
expressing gradable properties we exclude, e.g., form adjectives like rectangular, which
do not appear in adverbial function.) In Germanic languages, predicational adverbials
are typically realized by deadjectival adverbs. They appear in a wide variety of adverbial
usages. Typically, a single predicational can have at least two different usages, the exact
usage depending on its lexical semantics; cf. Ernst (1984, 2002). One example is given
in (11).
(11) a. Rudely, Claire greeted the queen.

b. Claire greeted the queen rudely.
In (11a) it is judged as rude that Claire greeted the queen, regardless of how she greeted
her; rudely serves as a subject-oriented adverbial here. In (11b), in contrast, what is quali-
fied as rude is not the very fact of greeting the queen, but the specific way in which Claire
greeted her; here rudely serves as a manner adverbial.
The most basic division in providing a further semantic subclassification for predi-
cational adverbials is that between sentence adverbials and verb-related adverbials
(sometimes also termed “higher” and “lower” adverbials). Sentence adverbials have
a hierarchically high attachment site; they stand in a relation to or combine with the
overall proposition expressed by the rest of the sentence without the adverbial (= the
sentential base). Verb-related adverbials have a lower attachment site within the VP and
are more closely connected to the verbal referent.
Some sort of distinction between sentence adverbials vs. verb-related adverbials along
the lines sketched above can be found in almost any semantic classification of adver-
bials, although details and further subdivisions may differ to some extent. The subdivi-
sion developed in the following draws on previous classifications, especially by Bartsch
(1972/1976), Jackendoff (1972), Bellert (1977), Ernst (1984, 2002), and Parsons (1990).
Each subclass will first be introduced on intuitive grounds and, if available, by some char-
acteristic paraphrases that are indicative of their underlying semantics. Afterwards, each
subclass will be characterized in terms of opacity, veridicality and further semantic and
inferential properties. (For a critical discussion of paraphrases, cf. e.g. Jackendoff (1972:
52) and Ernst (1984), for a very elaborate system of paraphrases, cf. Bartsch (1972).)
2.1.1. Sentence adverbials

Sentence adverbials can be further subdivided into subject-oriented adverbials, speaker-
oriented adverbials and domain adverbials.
Subject-oriented adverbials
The term goes back to Jackendoff’s (1972) “subject-oriented adverbs”. Subject-oriented
adverbials assign a specific property to the agent, based on the action as described by the
proposition expressed by the sentential base, cf. (12).
(12) Peter arrogantly/idiotically put his love letters on the net.
In (12), the speaker judges Peter to be arrogant/idiotic, basing his judgement on

Peter’s action of putting his love letters on the net. Sentences containing subject-oriented
adverbials allow paraphrases analogous to the one given in (13) for sentence (12).
(13) It was arrogant/idiotic of Peter to put his love letters on the net.
Subject-oriented adverbials are veridical and they have scope over negation: (14a)
entails (14b).
(14) a. Peter arrogantly did not answer my phone call.

b. Peter did not answer my phone call.
Finally, subject-oriented adverbials appear to be anomalous in questions, cf. (15).
(15) ?Did Peter arrogantly not answer my phone call?
Bellert (1977) relates this behavior to the general observation that we cannot ask a
question and assert a proposition in one and the same sentence. As Wyner (1994: 28ff)
and Geuder (2000: 165ff) point out, subject-oriented adverbials do not create opaque
contexts.
Speaker-oriented adverbials
Speaker-oriented adverbials provide a commentary by the speaker on the proposi-
tion expressed by the sentential base. They allow further subdivision into speech-act
adverbials, epistemic adverbials, and evaluative adverbials.
Speech-act adverbials characterize the speaker’s attitude towards the content (16a) or
the form (16b) of what s/he is saying; cf. Mittwoch (1977).
(16) a. Honestly/frankly, I have no idea what you’re talking about.

b. Briefly/roughly, Peter did not manage to convince her.
In declaratives, speech-act adverbials allow the addition of the participle speaking

without change in meaning, i.e. Honestly speaking, ... Furthermore, they can appear in
explicit perfomative utterances, e.g. I sincerely apologize.
Epistemic adverbials express the speaker’s expectation with regard to the truth of
the sentential base; cf. (17a). They can be paraphrased according to the pattern given
in (17b). (Note that maybe is special in that it is not gradable, but shares the general
characteristics of the other predicationals used here.)
(17) a. Maybe/probably/surely Mary is still alive.

b. It is maybe/probably/surely true that Mary is still alive.
Epistemic adverbials are often referred to as “epistemic modals”, contrasting with

alethic and deontic modals; cf., e.g., Parsons (1990: 62f) on epistemic vs. alethic modals
and Bonami, Godard & Kampers-Manhe (2004) on epistemic vs. alethic and deontic
interpretations of modals. An example for an alethic modal is the usage of necessarily in,
e.g., Two and two is necessarily four; deontic modals refer to rule or law based knowledge
as, e.g., In the USA, the president is necessarily the commander in chief; cf. article 58 (Hac-
quard) Modality. Epistemic adverbials cannot be directly negated (18a) nor do they have
negative counterparts (18b), and they are nonveridical (18c).
(18) a. *Matthew is not probably dead.

b. *Matthew is improbably dead.
c. Matthew is probably dead. ↛ Matthew is dead.
All three types of modals create opaque contexts for both subject and complement
positions, cf. the pattern for necessarily in (19) and (20).
(19) a. Necessarily, nine is an odd number.

b. The number of planets is nine.
c. ↛ Necessarily, the number of planets is an odd number.
(20) a. Necessarily, nine is an odd number.

b. Nine is a lucky number.
c. ↛ Necessarily, nine is a lucky number.
Evaluative adverbials express the opinion of the speaker with regard to the state of
affairs expressed by the rest of the sentence, cf. (21).
(21) Fortunately/surprisingly, Peter is back in Australia.
Paraphrases for evaluative adverbials follow the pattern of (23) for sentence (22).
(22) Fortunately/unfortunately, Peter is back in Australia.
(23) It is fortunate/unfortunate that Peter is back in Australia.
As the above example illustrates, evaluatives often come with negative counterparts,
although they usually cannot be negated analytically, cf. (24).
(24) a. Peter is fortunately back in Australia.

b. *Peter is not fortunately back in Australia.
They are veridical, and usually they cannot occur in hypothetical contexts, cf. (25). (See
Bellert (1977: 344f) for an explanation of why these two properties cooccur.)
(25) If firemen had (*unfortunately) not been available, my grandpa would maybe/
*fortunately have extinguished the fire himself.
Evaluatives are also anomalous in questions, cf. (26).
(26) *Is Peter fortunately back in Australia?
Because of the last two features, evaluative adverbials have recently been linked to posi-
tive polarity items; cf. Nilsen (2004) and Ernst (2007, 2009). With regard to opaque con-
texts, evaluatives behave similarly to subject-oriented adverbials, cf. Bonami, Godard &
Kampers-Manhe (2004).
Domain adverbials
Domain adverbials restrict the domain in which the proposition expressed by the rest of
the sentence is claimed to hold true; cf. Bellert (1977), McConnell-Ginet (1982), Bartsch
(1987), Ernst (2004).
(27) a. Emotionally Zardock is cold as ice.

b. Politically he is as good as dead.
c. Botanically, a tomato is a fruit.
Thus, (27a) says that the proposition expressed by Zardock is cold as ice is true when
the viewpoint on this proposition is restricted to the domain of emotions, but remains
neutral wrt. Zardock’s body temperature.
Domain adverbials do not appear to be veridical, cf. the pattern in (28).
(28) Deixis-wise, this sentence is intriguing.

↛ This sentence is intriguing.
The entailment failure in (28) is of a different nature than that with epistemic adverbials,
though. When dropping the domain adverbial, the sentence will still be evaluated from
a certain viewpoint. In this case the domain will be restricted to some default or contex-
tually salient value. That is, domain adverbials support an inferential pattern along the
lines of (28’). It is only because we cannot be sure that omitting the domain adverbial
will keep the implicitly involved domain constant that the inferential pattern in (28) does
not go through.
(28') Deixis-wise, this sentence is intriguing.

→ Wrt. some domain, this sentence is intriguing.
2.1.2. Verb-related adverbials

Verb-related adverbials have a lower attachment site within the VP and are more closely
connected to the verbal referent. Usually, at least mental-attitude adverbials, manner
adverbials, and degree adverbials are distinguished.
Mental-attitude adverbials
Mental-attitude adverbials describe the attitude of the agent with regard to the activity
described by the verbal predicate, cf. (29).
(29) Claire reluctantly/gladly went to school.
The adverbial reluctantly in (29) does not primarily describe the manner of going to
school, but Claire’s attitude towards going to school. It is only secondarily that this atti-
tude might also have an impact on Claire’s manner of going to school. Mental-attitude
adverbials can take scope over sentence negation, cf. (30).
(30) Martha gladly did not go to school.
However, in this case the agent does not have a certain attitude wrt. a negated proposi-
tion but wrt. the omission of a certain action, which is in turn an action. For instance, in
(30) Martha is glad about staying at home.
The mental-attitude adverbials in the above examples do not create opaque con-
texts. This is not a general property of mental-attitude adverbials, though. The mental-
attitude adverbial intentionally, for example, creates opaque contexts for the complement
position but not for the subject position; cf. (31), a classic example from Thomason &
Stalnaker (1973).
(31) Oedipus intentionally married Jocasta.

a. Oedipus is the son of Laius. → The son of Laius intentionally married Jocasta.
b. Jocasta is Oedipus’ mother. ↛ Oedipus intentionally married his mother.
Bonami, Godard & Kampers-Manhe (2004) label intentionally and similar items, like by
chance, “adverbs of attitude towards a state of affairs”.
Manner adverbials
Manner adverbials are used to specify the manner in which an eventuality or an action
unfolds; prototypical examples are given in (32).
(32) Klogman defended himself skillfully/intelligently/hectically.
Manner adverbials cannot take scope over sentence negation, cf. (33).
(33) Frankie does not run fast. ≠ Frankie does not run and he does so fast.
There is a straightforward semantic explanation for this behavior: Sentence negation

tells us that there is no eventuality of V-ing. Consequently there is no target available for
a potential manner modifier. Apparent counterexamples such as (34) are based on event
coercion. They require the interpolation of an event that can be plausibly associated
with the negated proposition; cf. articles 25 (de Swart) Mismatches and coercion and 34
(Maienborn) Event semantics.
(34) Klogman skillfully didn’t answer the question.
On a manner reading of skillfully in (34), what is skillful is some activity of Klogman

which allows him to uphold the state of not-answering the question, that is, he skillfully
dodged the question; cf. (Schäfer 2005: 161).
We will return to a more detailed discussion of manner adverbials in section 5.2.
Degree adverbials
Degree adverbials indicate the extent or intensity to which somebody does something;
cf. (35).
(35) Lochnan loves her very much/deeply.
Similarly to manner adverbials, degree adverbials cannot take scope over sentence
negation:
(36) Frankie does not love her very much. ≠ Frankie does not love her and he does so
very much.
Besides these three major subtypes there are further instances of verb-related adverbials
with a low attachment site such as the verb-related counterparts of domain adverbials,
the so-called method-oriented adverbials (cf. Schäfer 2005), which describe certain means
or methods of doing something, cf. (37).
(37) a. United Stated destroyed Switzerland economically.

b. The scientist classified the plants genetically.
c. They analyzed the data linguistically.
Some verb-related predicational adverbials may deviate from the standard behavior of
predicationals in non-trivial ways. Thus, halfway in (38) is neither veridical nor gradable.
(38) The door is halfway closed. ↛ The door is closed.
2.2. Participant-oriented adverbials

Participant-oriented adverbials—or circumstantials, as they are also called—are pre-
dominantly realized through prepositional phrases. They introduce a new participant that
takes part in the eventuality described by the verb. On a Neo-Davidsonian view, they are
linked to the verb’s eventuality argument through a thematic role just like standard agent
or patient arguments; cf. articles 18 (Davis) Thematic roles and 34 (Maienborn) Event
semantics. Sentence (39a), e.g., has two participant-oriented adverbials in the garage and
with a knife, which specify the place and the instrument role of the event. A standard
Neo-Davidsonian logical form is given in (39b); cf. section 4.3. for details.
(39) a. Peter opened the box with a knife in the garage.

b. ∃e [open (e) & agent (e, peter) & patient (e, the box) & location (e, the garage)
& instr (e, a knife)]
Just as predicationals, participant-oriented adverbials can have different uses. Following

the terminology in Maienborn (2001), we distinguish between event-related adverbials,
which restrict the verb’s eventuality argument, and frame adverbials, which set a frame
for the overall proposition; cf. the different meaning contributions of the locative,
temporal, and instrumental phrase in (40) vs. (41):
(40) a. We met Jürgen Klinsmann in the USA.

b. The Queen visited Jamestown in 1957.
c. Siri examined the diamond with a loupe.
(41) a. In the USA, resigned military officials are not frowned upon.
b. In 1957, moral integrity still had some value.
c. With a loupe, small fissures of a diamond become visible.
We will discuss the different uses of participant-related adverbials in more detail in

section 5.1.
2.3. Functional adverbials

Ernst’s (2002) last class, the so-called “functional adverbials” comprise a rather
heterogeneous set of adverbials including adverbial quantifiers as in (42a) as well as
discourse-anaphoric adverbials such as (42b).
(42) a. They often/never/usually carried out his orders.

b. They therefore/thus/notwithstanding became congenial companions.
We won’t discuss these adverbials any further here but refer the reader to articles 43
(Keenan) Quantifiers and 76 (Zimmermann) Discourse particles.
2.4. Adverbials and secondary predicates

Having laid out a semantic classification for adverbials we want to close this overview
with some remarks on the delineation of adverbials on the one hand and resultative
and depictive secondary predicates on the other hand; cf. also article 56 (Rothstein)
Secondary predicates.
Both resultatives and depictives introduce a secondary predicate into the sentence
that in a sense “lives on” the primary verbal predicate. This secondary predicate holds of
one of the verb’s arguments. That is, unlike verb-related adverbials, secondary predicates
do not qualify the verbal referent but one of its arguments. More specifically, depictives,
as in (43), express a secondary property of the subject or the object referent that holds
at least for the temporal duration of the verbal referent; cf. Rothstein’s (2003) notion of
time-participant-connectedness.
(43) Peter eats meat nude/raw.
Whenever a psychological adjective, i.e., an adjective denoting a particular state of mind,

is used, the distinction between mental-attitude adverbials and subject depictives is
blurred, especially in languages which do not use use different morphological forms to
differentiate between the adverbial use and the adjectival use as secondary predicate, cf.
the German example in (44).
(44) Gudrun ist traurig nach Hause gegangen.

Gudrun has sad/sadly to home gone
‘Gudrun went home sad/sadly.’
Geuder (2000) attempts to tease these different usages apart and contains a detailed
discussion of the English data. Himmelmann & Schulze-Berndt (2005) take a wide range
of typological data into account, showing that across languages there is considerable
variation in how depictives are encoded.
As for resultatives, they introduce a secondary predicate into the sentence that holds
true of one of the verb’s arguments as a result of the event expressed by the main predi-
cate; cf. e.g. (45), which expresses that the tulips became flat as a result of the gardener
watering them.
(45) The gardener watered the tulips flat.
There is a vast literature on resultatives; cf. the references in article 56 (Rothstein) Sec-
ondary predicates. One particular topic of interest relating to adverbials are manner-
resultative ambiguities such as the one in (46); elegantly may have a manner reading
as in (46a) as well as a resultative reading as in (46b); cf., e.g., the discussion in Eckardt
(1998, 2003), Geuder (2000), Dölling (2003). (Note that the resultative interpretation of
(46) involves a so-called “implicit resultative” (Schäfer 2005): Rather than predicating
over one of the verb’s overtly expressed arguments the secondary predicate holds for an
implicit argument, viz. Judith’s dress.)
(46) Judith dresses elegantly.

a. The way in which Judith dresses is elegant.
b. Judith dresses, so that as a result, her dress is elegant.
The manner and the resultative reading in (46) are conceptually easily distinguish-
able, because there isn’t any connection between the way one dresses and the result of
dressing. Yet, such a clear-cut distinction between manner and resultative readings is not
always possible; cf. the sentences in (47).
(47) a. Arndt fixed the chair perfectly.

b. Sarah grows roses marvelously.
The manner of fixing a chair or growing roses can only be qualified as perfect or mar-
velous if the result is of a comparably high quality and vice versa. If the result of, e.g.,
Arndt’s fixing the chair is perfect, then the way he did it must have been perfect, too. So,
manner and resultative readings cannot be completely disentangled in these cases; cf.
Quirk et al.’s (1985: 560) notion of blends.
2.5. Summary
The following Fig. 54.1 provides an overview of the adverbial subclasses that were
introduced in this section.
Fig. 54.1: Semantic classification of adverbials
3. Adverbials at the syntax/semantics interface

The semantic interpretation of an adverbial correlates to some degree with its syntactic
position. Jackendoff (1972) was the first to discuss this point in some detail; he distin-
guished three basic positions for adverbials in English: initial position, final position
without an intervening pause, and auxiliary position (i.e. between the subject and the
main verb). For illustration, consider English -ly adverbs. Some -ly adverbs can occur in
all three positions. But English also has -ly adverbs which can occur only in the initial and
aux positions along with -ly adverbs that occur only in the aux and final positions; cf. (48).
(48) a. (Frequently) Horatio has (frequently) lost his mind (frequently).

b. (Evidently/probably) Horatio (evidently/probably) lost his mind
*(evidently/probably).
c. *(Completely/easily) Stanly (completely/easily) ate his wheaties
(completely/easily).
Jackendoff argues that the different distributional patterns can also be distinguished
on semantic grounds, e.g. the adverbials showing the pattern in (48b) are speaker- or
subject-oriented, whereas manner adverbials show the pattern in (48c).
In the last decade, the correlation between syntactic position and semantic interpreta-
tion of adverbials has received considerable attention. Two main strands of thought can
be distinguished: an entirely syntax-driven one (represented by Cinque 1999), and one
based on semantic scope (represented by Ernst 1998, 1999, 2002 and Haider 1998, 2000).
Cinque (1999) has made an influential proposal to explain the order of adverb(ial)s
in purely syntactic terms, by assuming a universal hierarchy of functional heads that
encodes the hierarchy of adverbials. Adverbials are integrated as specifiers, each one
having a designated specifier position; cf. Alexiadou (1997) and Laenzlinger (1998) for
similar proposals and see also Alexiadou (2004a, b) for a recent overview. (Note that
Cinque (1999: 28ff) excludes participant-oriented adverbials—“circumstantials” in his
terms—from his adverb hierarchy because he considers them to lack a rigid ordering,
suggesting that they should be treated completely separately. Alexiadou (1997) and
Laenzlinger (1998) conceive of the universal adverb hierarchy as also including specifier
positions for circumstantials.)
Cinque’s purely syntactic account has been criticized by, e.g., Ernst (1998, 1999, 2002)
and Haider (1998, 2000) for leading to an unnecessary proliferation of functional heads
which duplicate underlying semantically motivated distinctions; see also Shaer (2003).
Ernst and Haider argue instead that the ordering restrictions on adverbials have no
genuine syntactic sources but can be derived from independent semantic properties.
According to this view, the syntax does not specify explicit attachment sites for (non-
subcategorized) adverbials but allows them to be adjoined wherever this is not explicitly
forbidden. The distribution of adverbials is accounted for by an interface condition map-
ping syntactic c-command domains onto semantic domains. Haider (1998, 2000) distin-
guishes three semantic domains: proposition > event > process/state. Ernst (1998, 2002)
assumes a richer hierarchy: speech act > fact > proposition > event > specified event.
Once the mapping procedure reaches a higher semantic domain, modifiers that address
the lower domain are ruled out.
The difference between the two approaches can be seen when looking at the sentence
pair in (49).
(49) a. Marie probably cleverly found a good solution.

b. *Marie cleverly probably found a good solution.
On Cinque’s account the ordering in (49a) is syntactically well-formed because this

reflects the assumed order of the relevant functional heads, whereas (49b) does not.
Ernst and Haider, on the other hand, argue that (49a) is fine, because cleverly selects
for events first, and probably, which requires an object of the higher semantic domain
proposition, is applied afterwards. When probably is applied first, as in (49b), the result
is of type proposition, which does not fit with cleverly anymore. Thus, on Ernst’s and
Haider’s account (49b) is semantically ill-formed.
While Ernst’s and Haider’s outline of a semantic explanation of the distributional
facts can be considered a promising alternative to Cinque’s hard-wired syntactic codi-
fication, many of its details remain to be worked out. For instance, as Frey (2003: 201ff)
points out, in Haider’s approach ordering restrictions are only assumed to hold between
adverbials. The placement of adverbials is not expected to be sensitive to the position
of arguments. Yet, as Frey (2003) shows, adverbials are not only ordered with respect to
each other but also with respect to the arguments of the verb. Furthermore, Frey (2003)
argues that on Ernst’s and Haider’s account adverbials shouldn’t be able to move around
but only appear in base-generated positions, otherwise they would be uninterpretable.
This doesn’t fit with the facts either in a scrambling language like German or in English;
cf. the discussion in Frey (2003) and see also the argumentation in Maienborn (2001) for
different base positions for locative adverbials.
With these observations in mind, Frey (2003) develops a compromise between a rigid
syntactic solution and a semantic scope approach by assuming five broader classes of
adverbials, each of which is assigned a syntactic base region defined by characteristic
structural requirements. Adverbials are freely base generated within the limits of their
characteristic region and they are allowed to move. (50) lists Frey’s five adverbial classes
and their syntactic positioning restrictions wrt. each other and wrt. the verb’s arguments
in terms of c-command (‘>’); cf. Frey (2003: 202f).
(50) Adverbial classes and base positions according to Frey (2003):

sentence adverbials > frame and domain adverbials > event-external adverbials
(e.g. causals) > highest ranked argument > event-internal adverbials (e.g. locatives,
instrumentals) > (internal arguments) > process-related adverbials (e.g. manner) >
verb
Frey’s proposal has been taken up, further elaborated and/or challenged by numerous
authors; cf. e.g. the articles in Lang, Maienborn & Fabricius-Hansen (2003).
A last complication at the syntax/semantics interface that should be mentioned here
involves parenthetical adverbials, that is, adverbials that are prosodically marked as
standing outside the regular syntactic structure. In English and German, these occur-
rences of adverbials appear with so-called comma-intonation, reflecting the corre-
sponding use of commata in writing. When adverbials are not integrated into a sentence,
they can appear in many more positions than when they are integrated, cf. (51a) vs. (51b).
(51) a. Peter obviously never came back home.

b. (Obviously,) Peter (, obviously,) never came back home (, obviously).
How these parentheticals are treated syntactically is not entirely clear. Their semantic
contribution often corresponds to at least one of the regular, integrated, usages, and
there tend to be preferences for a particular use specific to a given parenthetical posi-
tion; see Bonami, Godard & Kampers-Manhe (2004), Haegeman, Shaer & Frey (2009)
for more discussion and Shaer (2003, 2009) for a semantic analysis based on Haegeman’s
(1991/2009) orphans-approach.
4. Theoretical approaches
The foremost problem in dealing with adverbials in formal semantics is that there is
no natural place for them in the standard functor/argument set up. Neither do (non-
subcategorized) adverbials behave syntactically or semantically as “passive” arguments,
that are required by other categories and assigned to fixed positions, nor are they “active”
functors, opening up specific argument requirements and assigning structural positions.
We have to accept that standard formal semantics was not invented with adverbials in
mind. This makes them a particularly challenging subject for formal semantic accounts.
This section discusses three classical formal semantic treatments of adverbials, all of
which propose different ways of accounting for and reconciling the semantics of adver-
bials with some basic functor/argument account. These are (a) the operator approach
most prominently advocated by Thomason & Stalnaker (1973), (b) McConnell-Ginet’s
(1982) argument approach, and (c) the predicate approach, whose breakthrough came
with the spread of Davidsonian event semantics (Davidson 1967). (The order here does
not so much reflect the original publication history but rather the order of influence on
the linguistic community.)
4.1. The operator approach

The most influential text on adverbials as operators is Thomason & Stalnaker (1973),
but cf. also Montague (1970), Clark (1970), Parsons (1972), Cresswell (1973), and Kamp
(1975). Within this framework, adverbials are analyzed as endotypical functors. That is,
they are functors that, when applied to some argument of a certain logical type, yield a
result of the same type. This accounts for the typical iterability of adverbials: Since they
do not change the logical type of their environment they may be iterated.
Within this framework, Thomason & Stalnaker (1973) strive to account for the differ-
ences between adverbials like slowly and intentionally on the one hand and necessarily
on the other hand. They analyze the former as predicate modifiers and the latter as a sen-
tence modifier. One important difference between the two types of modifiers lies in their
behavior with regard to opaque contexts. The epistemic adverbial necessarily gives rise
to opaque contexts everywhere in a sentence, whereas intentionally only creates opaque
contexts for the object position; see section 2.1.1.
Thomason & Stalnaker (1973) account for this difference by analyzing sentence mod-
ifiers as functions from sentence intensions to sentence intensions, that is, necessarily is
of type <<s,t>,<s,t>>, and sentence (52a) can be represented as (52b), where the caret is
used to indicate the intension.
(52) a. Necessarily, nine is odd.

b. necessarily ^[odd (nine)]
Under this analysis the opaqueness effects are accounted for straightforwardly, because
sentence modifiers apply to sentence intensions.
In contrast, predicate modifiers map the intensions of one-place predicates into inten-
sions of one-place predicates. The restriction to one-place predicates means that, in
the case of transitive verbs, predicate modifiers are applied after the direct object has
combined with the verb, but before the verb combines with the subject, cf. (53).
(53) a. Oedipus intentionally married Jocasta.

b. intentionally ^[λx [marry (x, jocasta)]](oedipus)
This account correctly predicts that opacity arises with regard to the object position but
not with regard to the subject position. (Note that λ-conversion into an intension is not
possible here.) The opacity pattern exhibited by intentionally is thus elegantly accounted
for. Other adverbs like, e.g., slowly are treated in a similar way as intentionally, although
here we do not find parallel opacity effects, cf. (54).
(54) a. Renate slowly repaired the broken toy.

b. Renate is the director of the German Department.
→ The director of the German Department slowly repaired the broken toy.
c. The broken toy is my puppet. → Renate slowly repaired my puppet.
No theory-internal explanation is available for these patterns. Note, however, that slowly
cannot operate on predicate extensions, either, because this would lead to yet other
unwanted consequences; cf. e.g. McConnell-Ginet (1982: 162f): Given a scenario with co-
extensional dancers and singers, that is, all individuals who are singing are also dancing
and vice versa, there would be no way of distinguishing, say, the slow dancers from the
slow singers (due to Leibniz’ Law).
One of the major motivations for the operator approach, besides accounting for the
opacity effects, was a proper treatment of scope effects. A classical problem concerning
the scope of adverbials is illustrated by the sentence in (55) taken from Parsons (1972).
(55) John painstakingly wrote illegibly.
Parsons (1972: 131) argues that the correct interpretation of (55) requires that “the illeg-
ibility of the writing was at least one of the things John was taking pains to do”. That
is, painstakingly clearly has scope over illegibly. In the operator approach, this is pre-
dicted, because in the course of forming the complex predicate, the syntactically higher
adverbial is applied last, yielding (56).
(56) painstakingly [illegibly ^[λx [write (x)]]](john)
The second classical scope problem is discussed by Thomason & Stalnaker (1973) and
concerns the different readings available for (56a/b).
(57) a. Sam carefully sliced all the bagels.

b. Sam sliced all the bagels carefully.
While the exact reading differences for (57a/b) are somewhat subtle (cf. the discussion
in Eckardt 1998: 8f), they become more obvious if carefully is replaced, e.g., with quickly,
where quickly sliced all the bagels is preferably interpreted as meaning that the overall
time it took Sam to slice all the bagels was short, while sliced all the bagels quickly does
not tell us anything about the overall amount of time, but only gives the time span for
each individual slicing. Thomason & Stalnaker formalize this difference by having the
quantifier within the complex predicate for (57a), but letting it have widest scope for
(57b), see the formalizations in (58), where x is taken to range over bagels.
(58) a. carefully ^[λy ∀x [slice (y, x)]](sam)

b. ∀x [carefully ^[λy slice (y, x)](sam)]
The operator approach is usually also chosen to treat non-intersective adjectives in

attributive modification as, e.g., former in the former alcoholic.
4.2. The argument approach

An alternative to the operator approach is presented in McConnell-Ginet (1982).
McConnell-Ginet’s article discusses sentence adverbials as well as verb-related adver-
bials (her Ad-Verbs). Here, we will only focus on the latter. As mentioned above,
McConnell-Ginet shows that an extensional operator approach will not work for adverbs
like slowly. Furthermore, she argues that an intensional solution lacks psychological
plausibility and therefore isn’t adequate, either. Her own account draws on the observa-
tion that some manner adverbials are obligatory in a similar way as direct objects; see
the discussion of the sample sentences (2) in section 1.1. McConnell-Ginet goes on to
argue that verb-related adverbials in general should be treated as arguments of the verb.
According to this view, verbs have a latent potential of being further specified wrt. cer-
tain dimensions. What adverbials do is activate this potential and fill in a corresponding
value. For instance, the verb to run has a latent argument slot for speed, which may be
activated and filled in by an adverbial such as quickly. A simplified representation of
(59a) along these lines is given in (59b).
(59) a. Fritz runs quickly.

b. run (fritz, quickly)
In order to derive this representation, McConnell-Ginet introduces the operation

of verb-augmentation, by which additional argument slots can be made available when-
ever needed. Treating adverbials as arguments is particularly appealing in the case of
non-optional adverbials, e.g. for verbs like behave in (60).
(60) The kids behaved admirably.
McConell-Ginet’s approach distinguishes subcategorized and optional adverbials

only by the mode of integration. While subcategorized adverbials already have an argu-
ment slot available in the lexical entry of the verb, optional adverbials trigger verb-
augmentation. After this operation has taken place, the two types of adverbials are no
longer distinguishable. In addition, verb-augmentation does not distinguish different
types of Ad-Verbal modifiers. See Landman (2000) and Marten (2002) for further
discussions of McConnell-Ginet’s original approach and possible extensions.
4.3. The predicate approach

In his seminal paper “The logical form of action sentences” published in 1967 the lan-
guage philosopher Donald Davidson argues for a new ontological category of events.
This proposal has proven to be exceptionally fruitful for linguistics paving the way for
simpler and more adequate analyses of a multitude of linguistic phenomena; cf. article 34
(Maienborn) Event semantics. Davidson (1967) argues that a sentence such as (61a) does
not express a mere relation between Jones and the toast but introduces a hidden event
argument, which stands for the proper event of buttering, thus yielding (61b) as a formal
representation for (61a).
(61) a. Jones buttered the toast.

b. ∃e [butter (jones, the toast, e)]
One of the main motivations of Davidson’s proposal was to provide a straightforward

analysis of adverbial modification. If verbs introduce a hidden event argument, then
(intersective) adverbial modifiers can be analyzed as simple first order predicates that
add information about this event. Thus, Davidsons’s famous sentence (62a) receives
a formal representation as in (62b), or—adopting the so-called Neo-Davidsonian
framework of, e.g. Higginbotham (1985, 2000) and Parsons (1990)—as in (62c).
(62) a. Jones buttered the toast in the bathroom with the knife at midnight.
b. ∃e [butter (jones, the toast, e) & in (e, the bathroom) & instr (e, the knife) &
at (e, midnight)]
c. ∃e [butter (e) & agent (e, jones) & patient (e, the toast) & in (e, the bathroom)
& instr (e, the knife) & at (e, midnight)]
While Davidson’s original proposal was confined to participant-oriented adverbials,

Parsons (1990) extends the scope of the Davidsonian approach to manner adverbials like
slowly. Event-based treatments of mental-attitude adverbials are discussed in Eckardt
(1998), Wyner (1998), and Geuder (2000).
Davidson’s analysis of adverbials has two major merits. First, it accounts for the typ-
ical entailment patterns that characterize (intersective) adverbials directly on the basis
of their semantic representation. That is, the entailments in (63a–d) follow from (63)
simply by virtue of the logical rule of simplification. (Due to this feature, Davidson’s
approach cannot (easily) handle non-veridical adverbials.)
(63) Jones buttered the toast in the bathroom at midnight.

a. Jones buttered the toast in the bathroom and Jones buttered the toast at
midnight.
b. Jones buttered the toast in the bathroom.
c. Jones buttered the toast at midnight.
d. Jones buttered the toast.
Furthermore, Davidson’s approach does not allow us to infer (63) from (63b) and
(63c), since the latter sentences might relate to different events—a feature dubbed non-
entailment by Katz (2008). Again, this captures the data correctly.
The second major merit of Davidson’s account is that it treats adverbial modifiers on
a par with adnominal modifiers, thereby acknowledging their fundamental similarities.
Both adverbial and standard attributive modifiers provide one-place predicates, the only
difference being whether these predicates are applied to a noun’s referential argument
or to the verbal event argument. More generally speaking, the Davidsonian predicate
approach makes a considerable step forward towards a truly compositional semantics
for adverbials by teasing apart lexical and combinatorial ingredients of their meaning
contribution. The lexical meaning of a manner expression such as loud or a locative such
as in the garden simply denotes a certain property as in (64), irrespective of whether
these expressions happen to be used as adnominal (65) or adverbial (66) modifiers
(or as subcategorized arguments or main predicates together with the copula); cf., e.g.,
Bierwisch (1988), Wunderlich (1991), Maienborn (2001). (But see section 5.2. for some
qualifications concerning an adequate representation for manner expressions.)
(64) a. loud: λx [loud (x)]

b. in the garden: λx [in (x, the garden)]
(65) a. loud girl: λy [girl (y) & loud (y)]

b. girl in the garden: λy [girl (y) & in (y, the garden)]
(66) a. sing loudly: λe [sing (e) & loud (e)]

b. sing in the garden: λe [sing (e) & in (e, the garden)]
Given their common lexical roots it comes as no surprise that adverbials such as fast and
slowly in (67) display the same kind of context-dependency as their adjectival counter-
parts, requiring the inclusion of comparison classes; cf., e.g., the degree-based analyses of
these adjectives in Bierwisch (1989) and Kennedy (2007).
(67) Compared to other swimmers, Sarah crossed the channel fast, but compared to
Hovercrafts, she crossed it slowly.
In summing up we should note that the three classical accounts of the semantics of
adverbials were originally proposed as alternatives to each other, although they differ
considerably in scope. For instance, a Davidsonian predicate approach is not particu-
larly well-suited for adverbials that create opaque contexts, whereas McConnell-Ginet’s
(1982) argument approach seems especially attractive for subcategorized adverbials.
Moreover, there is no principled incompatibility between using events on the one hand
and analyzing at least some adverbials as operators on predicates; cf. Eckardt (1998: 12f).
Given the wide acceptance of events and their multifaceted use in present-day semantic
theory, current accounts of adverbial semantics mostly rely on the use of events as formal
semantic objects in some way or another. On this basis more sophisticated and differenti-
ated analyses of adverbial classes are being developed that strive to account, e.g., for the
particular behavior of adverbials wrt. information structure (see especially Eckardt 2003
on this point) as well as to deal with the further challenges that adverbials still pose. Two
of them concerning compositionality and ontological issues will be discussed in the next
section.
5. Challenges to compositionality and ontology

5.1. Uncovering the compositional machinery
In the previous section we pointed out that a Davidsonian predicate approach to adver-
bials makes a considerable step forward in separating the lexical and the combinatorial
meaning components that interact in yielding the characteristic semantics of adverbials.
(68) repeated from (64b) above specifies the lexical meaning of a locative adverbial
for illustration. The standard combinatorics may be spelled out by a modification tem-
plate MOD as in (69). MOD takes a modifier and an expression to be modified and
yields a conjunction of predicates thus accounting for the fundamental insight of the
Davidsonian predicate approach; cf. also Heim & Kratzer (1998) for an alternative
solution.
(68) in the garden: λx [in (x, the garden)]
(69) MOD: λQ λP λx [P(x) & Q(x)]
Leaving details aside, the application of MOD to an adverbial and a verbal eventuality
predicate will guarantee that the adverbial is predicated of the verb’s event argument as
in (70).
(70) sing in the garden: λe [sing (e) & in (e, the garden)]
This gives us the desired result—at least for the standard conception of intersective
adverbials. Unfortunately, matters turn out to be more intricate upon closer inspection.
Using locatives as a test case, Maienborn (1996, 2001, 2003) shows that, in addition to sup-
plying a holistic predicate of the verb’s event argument, circumstantial adverbials may
take various further interpretations. More specifically, Maienborn distinguishes three
different usages of locative adverbials: as frame adverbial, as event-external adverbial,
or as event-internal adverbial; cf. (71a–c), respectively. Only the event-external variant
in (71b) follows the standard MOD pattern in (69) whereas the frame and the event-
internal variants appear to behave differently. Since it would be both implausible and
theoretically unattractive to trace these meaning differences back to a lexical ambiguity
of the respective locatives, they must emerge somehow in the course of composition.
(71) a. In Argentina, Maradona is still very popular.

b. Maradona signed the contract in Argentina.
c. Maradona signed the contract on the last page.
The first noticeable difference is that frame adverbials (which we already mentioned
in section 2.2, see the discussion of (40)–(41)) pattern with domain adverbials in being
non-veridical. Frame adverbials are not part of what is properly asserted but restrict the
speaker’s claim. Therefore, their omission does not preserve truth if the domain restric-
tions expressed through the frame adverbial do not pattern with the default domain
restrictions; cf. the discussion of (28) in section 2.1.1. By contrast, both event-external and
event-internal locatives are veridical:
(72) a. In Argentina, Maradona is still very popular.

↛ Maradona is still very popular.
b. Maradona signed the contract in Argentina.
→ Maradona signed the contract.
c. Maradona signed the contract on the last page.
→ Maradona signed the contract.
Secondly, frame and event-internal adverbials differ from event-external adverbials in

being semantically underspecified in crucial respects. A frame adverbial such as (73) may
receive several interpretations along the dimensions spelled out, e.g., in (73a–c).
(73) In Italy, Maradona was married.

a. When he was in Italy, Maradona was married.
b. According to the laws in Italy, Maradona was married.
c. According to the belief of the people in Italy, Maradona was married.
That is, one can only say that frame adverbials restrict the speaker’s claim, but which
dimension exactly is being restricted is left semantically underspecified. Basically the
same holds true for event-internal adverbials. Their common semantic contribution con-
sists in specifying some internal aspect of the verb’s event argument, whose exact role
is left semantically implicit and can only be determined when taking into account con-
ceptual knowledge about the respective event type. Take, e.g., (71c): The locative (in its
preferred, event-internal, reading) does not express a location for the overall event of
Maradona signing the contract—this would be the event-external reading—but only for
one of its parts, viz. Maradona’s signature (which, by the way, isn’t referred to overtly in
the sentence).
A particularly puzzling feature of frame and event-internal locatives that is related to
their semantic indeterminacy is that they may take on non-locative interpretations. More
specifically, frame adverbials may have a temporal reading (cf. the paraphrase (73a)),
whereas event-internal adverbials tend to allow additional instrumental or manner
readings; cf. (74).
(74) a. The cook prepared the chicken in a Marihuana sauce.

b. The bank robbers escaped on bicycles.
c. Paul is standing on his head.
The adverbial in (74a) specifies a particular mode of preparing the food. Thus, it makes
some sort of manner contribution. The adverbial in (74b) supplies information about the
means of transport that was used by the bank robbers. It could be replaced by an instru-
mental phrase like with the cab. In the case of (74c), one might even doubt whether the
original locative meaning of the preposition is still present at all. In this case, there should
be an entity that is located on Paul’s head. What could that sensibly be? (Note that it
can’t be the regular subject referent Paul, which would include the head as a proper
part. Maienborn (2003: 498ff) proposes a possible answer to this puzzle that is based
on the locative’s regular meaning. According to this solution it is Paul’s remaining body
(modulo his head) that is located on—and thus supported by—Paul’s head.)
Note that these supplementary, non-locative readings of frame and event-internal ad-
verbials are most appropriately queried by using the respective non-locative interrogatives:
(73’) a. When was Maradona married?
(74’) a. How did the cook prepare the chicken?

b. How / With what did the bank robbers escape?
c. How is Paul standing?
Standard event-external adverbials, on the other hand, always refer to the overall loca-
tion of the verb’s event argument. They do not share the ability of event-internal and
frame modifiers to convey additional non-locative information, and they can only be
questioned by a locative interrogative.
The challenge that circumstantial adverbials such as locatives pose to a formal seman-
tics of adverbs is, on the one hand, that there is good reason to assume that expressions
such as in Argentina or on the last page have a unique lexical meaning, i.e. they express
the property of some entity being located in a particular spatial location. On the other
hand, we have to account for the different readings of locatives and their characteristic
properties in terms of inferential behavior, semantic indeterminacy and the emergence
of supplementary non-locative interpretations.
In a nutshell, the solution proposed in Maienborn (1996, 2001, 2003) takes the fol-
lowing track. First, it is shown that there is a strict correlation between the position of a
locative adverbial and its interpretation. More specifically, the three types of locatives are
argued to have distinctive syntactic base positions, each corresponding to one of Frey’s
(2003) adverbial positions; see (50). Event-internal adverbials are base-generated at
the V-periphery, event-external adverbials are base-generated at the VP-periphery, and
frame adverbials have a high base-adjunction site within the C-Domain. These distinct
structural positions provide the key for a compositional account, since an adverbial will
be linked up with different target referents depending on its structural position. While
event-external adverbials are linked up to the verb’s event argument, event-internal and
frame adverbials are semantically underspecified in this respect. Event-internal adver-
bials are linked up to a referent that is related to the verb’s event argument, and frame
adverbials are linked up to a referent that is related to the topic of the sentence. The
identification of these target referents is shown to depend on discourse and world knowl-
edge. The non-locative readings of event-internal and frame adverbials are reconstructed
as a side effect of the pragmatic resolution of semantic indeterminacy; cf. also articles
24 (Egg) Semantic underspecification and 31 (Lang & Maienborn) Two-level Semantics.
Maienborn proposes a compositional account for these adverbials that is sensitive to the
observed structural and pragmatic influences while still preserving the basic insights of
the classical Davidsonian approach. To this end, the template MOD in (69) is replaced
by a more general variant MOD* in (75), whose application is regulated by the interface
condition in (76); cf. Maienborn (2003: 489).
(75) MOD*: λQ λP λx [P(x) & R (x, v) & Q(v)]
(76) Condition on the application of MOD*:

If MOD* is applied in a structural environment of categorial type X, then R =
part-of, otherwise (i.e. in an XP-environment) R is the identity function.
MOD* introduces a free variable v and a relational variable R. If applied in an XP-

environment, R is instantiated as identity, i.e. v is identified with the referential argument
of the modified expression, thus yielding the standard variant MOD. This is the case
with event-external adverbials. If MOD* is applied in an X-environment, R is instanti-
ated as part-of; cf. also Dölling (2003) for a formal account of the flexibility of adverbial
modification that is similar in spirit.
The relation part-of pairs entities with their integral constituents. In the case
of events, among these are the event’s participants. The result of applying MOD* to a
sentence with an event-internal adverbial such as (77a) is given in (77b).
(77) a. The bank robber escaped on the bicycle.

b. ∃e [escape (e) & theme (e, bank robber) & part-of (e, v) & on (v, bike)]
According to the semantic representation in (77b), an entity v which is involved in the

escaping event is located on the bicycle. This is as far as the compositional semantics
of event-internal adverbials takes us. The identification of v and its exact role in e can
only be spelled out at the conceptual level taking into account world knowledge, e.g.,
about extrinsic and intrinsic movement, the use of vehicles for extrinsic movement, spa-
tial prerequisites that need to be fulfilled in order for a vehicle to function properly, etc.
A simplified conceptual spell-out for (77b) is given in (77c); cf. Maienborn (2003: 490ff)
for details; see also article 31 (Lang & Maienborn) Two-level Semantics.
(77) c. ∃e [escape (e) & extrinsic-move (e) & theme (e, bank robber)
& instr (e, bike) & vehicle (bike) & support (bike, bank robber)
& on (bank robber, bike)]
This conceptual spell-out provides a plausible utterance meaning for sentence (77a). It
goes beyond the compositionally determined meaning in the following respects: (a) it
specifies that the escape was taken by extrinsic means (extrinsic-move). As a conse-
quence, (b) the bike is identified as the instrument of locomotion in the given event. This
in turn leads (c) to an instantiation of the free variable v by the discourse referent repre-
senting the bank robber and an identification of the part-of relation with the theme-role.
For other cases, as, e.g., (78) more conceptual inferencing will be required in order to
identify a suitable referent to which the event-internal locative applies.
(78) Paul tickled Maria on her neck.
That is, what turns out to be located on Maria’s neck in (78) could be, e.g., Paul’s hand
or maybe some feather he used for tickling Maria. Although not manifest at the linguis-
tic surface, such conceptually inferred units qualify as potential instantiations of the
compositionally introduced free variable v.
Maienborn (2001: §6) sketches how MOD* may also account for the semantics
of frame adverbials. Generalizing Klein’s (1994) notion of topic time, frame adverbials
can be seen as providing an underspecified restriction on an integral part of a topic
situation.
All in all Maienborn’s proposal suggests that the flexibility of adverbial modifica-
tion is the result of adverbials (a) having several potential structural integration sites in
combination with (b) being subject to a particular kind of semantic indeterminacy.
5.2. What are manners?

In section 2 above, we distinguished, among other things, between manner adverbials,
degree adverbials, and other adverbials, half-way being one of them. But already the
notion of manner adverbials is not very clearly defined, and is usually taken to com-
prise a rather large group of adverbials. Thus, all adverbs in (79) are typically considered
manner adverbials.
(79) a. Peter runs fast/slowly.

b. Marie sings loudly/quietly.
c. Kim dances beautifully/woodenly.
d. Claire solved the problem skillfully/intelligently.
All these adverbials can be questioned by How ...? They are all veridical, and they cannot
take scope over sentence negation nor do they create opaque contexts. Nevertheless, their
meaning contributions to the sentence are very different. This can be easily seen by looking
at the behavior of the adverbials in (79) with regard to standard paraphrases for manner
adverbials. Standard paraphrases like ... in a ADJ manner or The way X VERBs is ADJ
are not appropriate for all these items. They are perfectly applicable to (79c): Kim dances
beautifully/in a beautiful manner and The way Kim dances is beautiful are synonymous.
However, they do not fit for (79a/b): to run fast means that the speed of the running was
fast, not the manner. Similarly, to sing loudly means that the sound-volume of the singing
was loud, not the manner. Finally, (79d) seems to correspond to these paraphrases only on
one reading, according to which Claire reached the solution by a series of intelligent steps.
On a reading of (79d) according to which the solution arrived at is an intelligent one, the
paraphrases turn out to be inappropriate, and a classification of this reading as an resulta-
tive or a blend might be more fitting. A further difference between (79d) and the other
adverbials in (79a–c) is that it involves a direct relation to the subject: Roughly, the subject
appears as intelligent through the way of solving the problem or the kind of the solution s/
he provided. Obviously, assuming a plain analysis as one-place predicates over events for
the adverbials in (79) won’t suffice to account properly for all these peculiarities.
(80) a. Peter talked loudly.

b. ∃e [agent (e, peter) & talk (e) & loud (e)]
An analysis of (80a) along the lines of (80b) does not make explicit that the adverbial
specifies the sound-volume of the talking, i.e., that it is specifying one particular aspect
of the talking event. Another strange effect of a plain Neo-Davidsonian representation
is that the verb and the manner adverbial appear to be semantically on a par (both pro-
viding one-place predicates over events) while intuitively and syntactically, they are not.
One possible way toward a more elaborate theory of manner adverbials that helps
overcome some of these shortcomings consists in introducing manners as a further onto-
logical category in our formal language. This idea has recently been brought back into
the discussion by Piñón (2007, 2008). Its first, dismissive, discussion can be found in Fodor
(1972), whereas Dik (1975) was the first champion of this approach. The main idea is
simple enough: we need to be able to access the conceptual properties of the events intro-
duced by the verb in order to gain an adequate understanding of manner modification.
Thus, in order to capture the fact that loudly assigns the property loud to the sound-
volume of the talking, we need to retrieve the corresponding conceptual coordinate of
the talking event. Similarly, for fast, we need the conceptual coordinate for speed. What
kind of coordinate do we need for adverbs like beautifully and intelligently? The avail-
ability of the manner paraphrase for these adverbs shows us that we need a coordinate
that is more complex than those needed for loudly and fast and that it cannot be reduced
to what are essentially quite straightforward, monodimensional scales of the intuitively
clear concepts speed and sound-volume. For the sake of simplicity, we will assume that in
both cases the required coordinate is in fact a manner of the events in question, so that,
consequently, beautifully is predicated of the manner of dancing, and intelligently of the
manner of answering the question. A simplified illustration of a semantic representa-
tion for (79c) is given below, where the conceptual coordinate manner is linked to the
event argument via an underspecified relation R; cf. (81). (This corresponds to one of the
versions considered in Fodor 1972.)
(81) a. Peter danced beautifully.

b. ∃e [agent (e, peter) & dance (e) & ∃m [R (e, m) & beautiful (m)]]
While clearly pointing in the right direction, this approach obviously also raises many
intricate questions. While we cannot do justice to all of them here, it is helpful to briefly
consider the pros and cons of this approach.
As Piñón (2007) points out, one argument in favor of assuming manner as an onto-
logical entity is that it can be perceived, as evidenced by expressions such as (82); cf. the
discussion on perception reports as one of the main criteria for assuming the ontological
category of events in article 34 (Maienborn) Event semantics.
(82) I saw how Linda danced.
Furthermore, as Piñón (2007) argues, assuming manners also allows us to systematically

relate the in an X manner paraphrase to manner adverbials, since in both cases we have
predicates of manners. The head noun of the paraphrasing prepositional phrase refers
to a manner, and the attributive adjective predicates of this manner. In the very same
way, its adverbial counterpart predicates of the manner made available as a conceptual
coordinate of the event referred to by the verb.
Finally, this fine-grained analysis of manner modification can also be used to account
for otherwise unexplainable patterns, e.g., the different behavior of the adverb audibly
in (83a/b) discussed in Cresswell (1985: 186ff).
(83) a. Isolde audibly precedes/follows Jeremy.

b. Kiri sings/dances audibly.
As Cresswell points out, in the case of (83a), it can be some activity other than the pre-
ceding/following itself that causes the audibility, whereas in the case of (83b), what is
audible is the sound of the singing/dancing. This observation can be accounted for by
assuming that the conceptual structure of dancing/singing events differs from the con-
ceptual structure of preceding/following events in that only the former but not the latter
readily provide the corresponding sound-coordinate. The scope-taking usages of manner
adverbials discussed in section 4.1 can also be accounted for by resorting to an analysis
based on manners, cf. Piñón (2007) and Schäfer (2008) for two formal accounts.
Obvious objections to this approach concern matters of ontology: What exactly are
manners supposed to be, and what do we mean when we speak of coordinates of events?
Manners, speeds, and sound volumes are all ontologically dependent on the events intro-
duced by the verbs in the respective sentences, that is, they do not and cannot exist by
themselves. These ontologically dependent entities can be viewed as coordinates in the
conceptual structure of their host events. The exact nature and internal structure of these
coordinates is still an unanswered question, but Geuder’s (2006) discussion of manner
adverbs and their relation to conceptual dimensions is a promising starting point; cf. also
the notion of dossiers in article 16 (Bierwisch) Semantic features and primes.
Note that this analysis has some striking resemblance to the semantics for event-
internal adverbials proposed in the previous section; cf. the discussion of MOD* in (75).
That is, conceptually dependent units such as speed, sound-volume, or manner may be
made accessible for further specification via a semantically underspecified event relation.
Event-internal circumstantials and manner adverbials thus both enable and enforce
a closer look into the internal structure of events. Obviously, much remains to be done
in this area. Manner adverbials, despite their innocent appearance as being the paradig-
matic case for a textbook Davidsonian analysis, still turn out to pose many riddles that a
formal semantics for adverbials will have to solve.
6. Conclusion
Adverbials and their dedicated word class, the adverbs, comprise a heterogeneous set
of lexical and semantic classes and subclasses with very specific inferential and distribu-
tional properties. They are only loosely tied to the surrounding syntactic and semantic
structure, leaving much space for variation and adaptation. What the vast majority of
adverbs has in common is that they are non-subcategorized linguistic parasites: Wherever
they find a suitable integration site, they attach to it and supply additional and uncalled-
for information. Precisely because of this parasitic nature and their frappant flexibility,
adverbials constitute a challenge for linguistic theory, which, in turn, must account for
this flexible means of natural language expression in terms of a sufficiently rigid account
of their lexical, compositional, and conceptual semantics.
7. References
Alexiadou, Artemis 1997. Adverb Placement: A Case Study in Antisymmetric Syntax. Amsterdam:
Benjamins.
Alexiadou, Artemis 2004a. Adverbs across frameworks. Lingua 114, 677–682.
Alexiadou, Artemis (ed.) 2004b. Taking up the Gauntlet. Adverbs across Frameworks. Lingua 114,
677–682.
Austin, Jennifer, Stefan Engelberg & Gesa Rauh (eds.) 2004. Adverbials. The Interplay between
Meaning, Context, and Syntactic Structure. Amsterdam: Benjamins.
van der Auwera, Johan (ed.) 1998. Adverbial Constructions in the Languages of Europe. Berlin:
Mouton de Gruyter.
Bartsch, Renate 1972/1976. Adverbialsemantik. Die Konstitution logisch-semantischer Repräsen-
tationen von Adverbialkonstruktionen. Frankfurt/M. Athenäum. English Translation: The
Grammar of Adverbials. A Study in the Semantics and Syntax of Adverbial Constructions.
Amsterdam: North-Holland, 1976.
Bartsch, Renate 1987. The construction of properties under perspectives. Journal of Semantics
5, 293–320.
Bellert, Irena 1977. On semantic and distributional properties of sentential adverbs. Linguistic
Inquiry 8, 337–351.
Bierwisch, Manfred 1988. On the grammar of local prepositions. In: M. Bierwisch, W. Motsch & I.
Zimmermann (eds.). Syntax, Semantik und Lexikon. Berlin: Akademie Verlag, 1–65.
Bierwisch, Manfred 1989. The semantics of gradation. In: M. Bierwisch & E. Lang (eds.). Dimen-
sional Adjectives. Berlin: Springer, 71–261.
Bonami, Olivier, Danièle Godard & Brigitte Kampers-Manhe 2004. Adverb classification. In: F.
Corblin & H. de Swart (eds.). Handbook of French Semantics. Stanford, CA: CSLI Publications,
143–184.
Cinque, Guglielmo 1999. Adverbs and Functional Heads. A Cross-Linguistic Perspective. Oxford:
Oxford University Press.
Clark, Romane 1970. Concerning the logic of predicate modifiers. Noûs 4, 311–335.
Cresswell, Max J. 1973. Logics and Languages. London: Methuen.
Cresswell, Max J. 1985. Adverbial Modification. Interval Semantics and Its Rivals. Dordrecht: Reidel.
Davidson, Donald 1967. The logical form of action sentences. In: N. Rescher (ed.). The Logic of
Decision and Action. Pittsburgh, PA: University of Pittsburgh Press, 81–95. Reprinted in: D.
Davidson (ed.). Essays on Actions and Events. Oxford: Oxford University Press, 1980, 105–122.
Delfitto, Denis 2000. Adverbs and the syntax/semantics interface. Italian Journal of Linguistics
12, 13–53.
Dik, Simon C. 1975. The semantic representation of manner adverbials. In: A. Kraak (ed.). Linguis-
tics in the Netherlands 1972–1973. Assen: Van Gorcum, 96–121.
Dölling, Johannes 2003. Flexibility in adverbal modification: Reinterpretation as contextual enrich-
ment. In: E. Lang, C. Maienborn & C. Fabricius-Hansen (eds.). Modifying Adjuncts. Berlin: de
Gruyter, 511–552.
Eckardt, Regine 1998. Events, Adverbs, and Other Things. Issues in the Semantics of Manner
Adverbs. Tübingen: Niemeyer.
Eckardt, Regine 2003. Manner adverbs and information structure. Evidence from the adverbial
modification of verbs of creation. In: E. Lang, C. Maienborn & C. Fabricius-Hansen (eds.). Modi-
fying Adjuncts. Berlin: de Gruyter, 261–305.
Egg, Markus 2006. Anti-Ikonizität an der Syntax-Semantik-Schnittstelle. Zeitschrift für Sprachwis-
senschaft 25, 1–38.
Ernst, Thomas 1984. Towards an Integrated Theory of Adverb Position in English. Bloomington, IN:
Indiana University Linguistics Club.
Ernst, Thomas 1998. Scope based adjunct licensing. In: P. Tamanji & K. Kusumoto (eds.). Proceed-
ings of the Annual Meeting of the North East Linguistic Society (= NELS) 28. Amherst, MA:
GLSA, 127–142.
Ernst, Thomas 1999. Adjuncts, the universal base, and word order typology. In: P. Tamanji, M.
Hirotami & N. Hall (eds.). Proceedings of the Annual Meeting of the North East Linguistic
Society (= NELS) 29. Amherst, MA: GLSA, 209–223.
Ernst, Thomas 2002. The Syntax of Adjuncts. Cambridge: Cambridge University Press.
Ernst, Thomas 2004. Domain adverbs and the syntax of adjuncts. In: J. Austin, S. Engelberg &
G. Rauh (eds.). Adverbials. The Interplay between Meaning, Context, and Syntactic Structure.
Amsterdam: Benjamins, 103–129.
Ernst, Thomas 2007. On the role of semantics in a theory of adverb syntax. Lingua 117, 1008–
1033.
Ernst, Thomas 2009. Speaker-oriented adverbs. Natural Language and Linguistic Theory 27,
497–544.
Fodor, Jerry A. 1972. Troubles about actions. In: D. Davidson & G. Harman (eds.). Semantics of
Natural Language. Dordrecht: Reidel, 48–69.
Frey, Werner 2003. Syntactic conditions on adjunct classes. In: E. Lang, C. Maienborn &
C. Fabricius-Hansen (eds.). Modifying Adjuncts. Berlin: de Gruyter, 163–209.
Geuder, Wilhelm 2000. Oriented Adverbs. Issues in the Lexical Semantics of Event Adverbs.
Doctoral dissertation. University of Tübingen.
Geuder, Wilhelm 2006. Manner modification of states. In: Ch. Ebert & C. Endriss (eds.). Proceed-
ings of Sinn und Bedeutung (= SuB) 10. Berlin: ZAS, 111–124.
Giannakidou, Anastasia 1999. Affective dependencies. Linguistics & Philosophy 22, 367–
421.
Goldberg, Adele & Farrell Ackerman 2001. The pragmatics of obligatory adjuncts. Language
77, 798–814.
Haegeman, Liliane 1991/2009. Parenthetical adverbials: The radical orphanage approach. In:
S. Chiba et al. (eds.). Aspects of Modern English Linguistics: Papers presented to Masatomo
Ukaji on his 60th Birthday. Tokio: Kaitakusha, 232–254. Reprinted in: B. Shaer et al. (eds.). Dis-
located Elements in Discourse. Syntactic, Semantic, and Pragmatic Perspectives. New York: Rout-
ledge, 2009, 331–347.
Haegeman, Liliane, Benjamin Shaer & Werner Frey 2009. Postscript: Problems and solutions
for orphan analyses. In: B. Shaer et el. (eds.). Dislocated Elements in Discourse. Syntactic,
Semantic, and Pragmatic Perspectives. New York: Routledge, 348–365.
Haider, Hubert 1998. Adverbials at the syntax-semantics interface. In: H. Kamp & U. Reyle (eds.).
Tense and Aspect Now. Perspectives on Problems in the Theory of Tense and Aspect. Tübingen:
Niemeyer, 51–67.
Haider, Hubert 2000. Adverb placement: Convergence of structure and licensing. Theoretical Lin-
guistics 26, 95–134.
Heim, Irene & Angelika Kratzer 1998. Semantics in Generative Grammar. Oxford: Blackwell.
Higginbotham, James 1985. On semantics. Linguistic Inquiry 16, 547–593.
Higginbotham, James 2000. On events in linguistic semantics. In: J. Higginbotham, F. Pianesi & A.
Varzi (eds.). Speaking of Events. Oxford: Oxford University Press, 49–79.
Himmelmann, Nikolaus P. & Eva Schultze-Berndt 2005. Issues in the syntax and semantics of
participant-oriented adjuncts: An introduction. In: N. P. Himmelmann & E. Schultze-Berndt
(eds.). Secondary Predication and Adverbial Modification. The Typology of Depictives. Oxford:
Oxford University Press, 1–67.
Huddleston, Rodney & Geoffrey K. Pullum 2002. The Cambridge Grammar of the English Lan-
guage. Cambridge: Cambridge University Press.
Jackendoff, Ray 1972. Semantic Interpretation in Generative Grammar. Cambridge, MA: The MIT
Press.
Kamp, Hans 1975. Two theories about adjectives. In: E. L. Keenan (ed.). Formal Semantics of Nat-
ural Language. Cambridge: Cambridge University Press, 123–155.
Katz, Graham 2008. Manner modification of state verbs. In: L. McNally & Ch. Kennedy (eds.).
Adjectives and Adverbs. Syntax, Semantics, and Discourse. Oxford: Oxford University Press,
220–248.
Kennedy, Christopher 2007. Vagueness and grammar: The semantics of relative and absolute grad-
able predicates. Linguistics & Philosophy 30, 1–45.
Klein, Wolfgang 1994. Time in Language. London: Routledge.
Krifka, Manfred 1992. A compositional semantics for multiple focus construction. In: J. Jacobs (ed.).
Informationsstruktur und Grammatik. Opladen: Westdeutscher Verlag, 17–53.
Laenzlinger, Christopher 1998. Comparative Studies in Word Order Variation: Adverbs, Pronouns
and Clause Structure in Romance and Germanic. Amsterdam: Benjamins.
Landman, Fred 2000. Events and Plurality. The Jerusalem Lectures. Dordrecht: Kluwer.
Lang, Ewald, Claudia Maienborn & Cathrine Fabricius-Hansen (eds.) 2003. Modifying Adjuncts.
Berlin: de Gruyter.
Larson, Richard K. 1998. Events and modification in nominals. In: D. Strolovitch & A. Lawson
(eds.). Proceedings from Semantics and Linguistic Theory (= SALT) VIII. Ithaca, NY: Cornell
University Press, 145–168.
Maienborn, Claudia 1996. Situation und Lokation. Die Bedeutung lokaler Adjunkte von Verbalpro-
jektionen. Tübingen: Stauffenburg.
Maienborn, Claudia 2001. On the position and interpretation of locative modifiers. Natural Lan-
guage Semantics 9, 191–240.
Maienborn, Claudia 2003. Event-internal adverbials: Semantic underspecification and conceptual
interpretation. In: E. Lang, C. Maienborn & C. Fabricius-Hansen (eds.). Modifying Adjuncts.
Berlin: de Gruyter, 475–509.
Marten, Lutz 2002. At the Syntax-pragmatics Interface: Verbal Underspecification and Concept For-
mation in Dynamic Syntax. Oxford: Oxford University Press.
McConnell-Ginet, Sally 1982. Adverbs and logical form. A linguistically realistic theory. Language
58, 144–184.
McNally, Louise & Christopher Kennedy (eds.) 2008. Adjectives and Adverbs. Syntax, Semantics,
and Discourse. Oxford: Oxford University Press.
Mittwoch, Anita 1977. How to refer to one’s own words: Speech-act modifying adverbials and the
performative analysis. Journal of Linguistics 13, 177–189.
Montague, Richard 1970. English as a formal language. In: B. Visentini (ed.). Linguaggi nella
Società e nella Tecnica. Milan: Edizioni di Comunità, 189–223. Reprinted in: R. H. Thomason
(ed.). Formal Philosophy. Selected Papers of Richard Montague. New Haven, CT: Yale Univer-
sity Press, 188–221.
Morzycki, Marcin 2008. Adverbial modification of adjectives: Evaluatives and a little beyond. In: J.
Dölling, T. Heyde-Zybatow & M. Schäfer (eds.). Event Structures in Linguistic Form and Inter-
pretation. Berlin: de Gruyter, 103–126.
Nilsen, Øystein 2004. Domains for adverbs. Lingua 114, 809–847.
Parsons, Terence 1972. Some problems concerning the logic of grammatical modifiers. In: D.
Davidson & G. Harman (eds.). Semantics of Natural Language. Dordrecht: Reidel, 127–141.
Parsons, Terence 1990. Events in the Semantics of English. A Study in Subatomic Semantics.
Cambridge, MA: The MIT Press.
Piñón, Christopher 2007. Manner adverbs and manners. Handout, 7. Ereignissemantik-Konferenz,
Tübingen, December 20, 2007. http://pinon.sdf-eu.org/covers/mam.html, August 9, 2011.
Piñón, Christopher 2008. From properties to manners: A historical line of thought about manner
adverbs. In: A. Lobke et al. (eds.). Papers of the Linguistic Society of Belgium, vol. 3, 2008. http://
pinon.sdf-en.org/covers/pm.html, August 9, 2011.
Quirk, Randolph, Sidney Greenbaum, Geoffrey Leech & Jan Svartvik 1985. A Comprehensive
Grammar of the English Language. Harlow: Longman.
Rawlins, Kyle 2008. Unifying illegally. In: J. Dölling, T. Heyde-Zybatow & M. Schäfer (eds.). Event
Structures in Linguistic Form and Interpretation. Berlin: de Gruyter, 81–102.
Rothstein, Susan 2003. Secondary predication and aspectual structure. In: E. Lang, C. Maienborn &
C. Fabricius-Hansen (eds.). Modifying Adjuncts. Berlin: de Gruyter, 553–590.
Sasse, Hans-Jürgen 1993. Syntactic categories and subcategories. In: J. Jacobs, A. v. Stechow, W.
Sternefeld & T. Vennemann (eds.). Syntax– Ein internationales Handbuch zeitgenössischer
Forschung (HSK 9). Berlin: de Gruyter, 646–686.
Schäfer, Martin 2005. German Adverbial Adjectives: Syntactic Position and Semantic Interpretation.
Doctoral dissertation. University of Leipzig.
Schäfer, Martin 2008. Resolving scope in manner modification. In: O. Bonami & P. Cabredo
Hofherr (eds.). Empirical Issues in Syntax and Semantics 7. Paris: CSSP, 351–372.
Shaer, Benjamin 2003. “Manner” adverbs and the association theory: Some problems and solutions.
In: E. Lang, C. Maienborn & C. Fabricius-Hansen (eds.). Modifying Adjuncts. Berlin: de Gruyter,
211–259.
Shaer, Benjamin 2009. German and English left-peripheral elements and the “orphan” analysis.
In: B. Shaer et al. (eds.). Dislocated Elements in Discourse. Syntactic, Semantic, and Pragmatic
Perspectives. New York: Routledge, 366–397.
Thomason, Richard H. & Robert C. Stalnaker 1973. A semantic theory of adverbs. Linguistic
Inquiry 4, 195–220.
Wunderlich, Dieter 1984. Zur Syntax der Präpositionalphrasen im Deutschen. Zeitschrift für
Sprachwissenschaft 3, 65–99.
Wunderlich, Dieter 1991. How do prepositional phrases fit into compositional syntax and seman-
tics? Linguistics 29, 591–621.
Wyner, Adam Zachary 1994. Boolean Event Lattices and Thematic Roles in the Syntax and Seman-
tics of Adverbial Modification. Ph.D. dissertation. Cornell University, Ithaca, NY.
Wyner, Adam Zachary 1998. Subject-oriented adverbs are thematically dependent. In: S. Rothstein
(ed.). Events in Grammar. Dordrecht: Kluwer, 333–348.
Claudia Maienborn, Tübingen (Germany) and

Martin Schäfer, Jena (Germany)
55. Adverbial clauses

1. Introduction
2. Temporal clauses
3. Modal clauses
4. Instrumental and free adjunct clauses
5. Conclusions
6. References
Abstract
Adverbial clauses are subordinate clauses that modify their superordinate clauses. This
modification can occur at various levels (such as verb phrase, tense phrase, mood phrase)
and in various dimensions (such as times and worlds) and ways. These variations give
rise to a categorization of adverbial clauses (temporal, modal, ...) and a subcategoriza-
tion according to a range of relations within these dimensions, depending on the sub-
junction. Thus within the modal category it is customary to distinguish between causal,
conditional, purpose, result, and concessive clauses. Sometimes the subjunction does not
seem to encode much meaning of its own and the clause acts more like a relative clause,
modifying a quantificational adverb or a modal, or specifying an underspecified predicate;
sometimes, when there is no subjunction (“free” adjunct clauses), the contribution of the
clause is underspecified.
1. Introduction
Adverbial clauses are a proper subclass of the class of all adverbials. To a considerable
extent, this subclass relation distributes over the major semantic categories of adverbials
commonly identified (see article 54 (Maienborn & Schäfer) Adverbs and adverbials).
Thus in the temporal category, there are closely comparable clausal and nonclausal
adverbials, e.g.: “since Benitez arrived at Anfield” - “since June 2004”. Locative adver-
bials and manner adverbials tend to be nonclausal, but instrumental adverbials can be
both: “by hammering it” - “with a hammer”. In the modal category, clausal adverbials
predominate; there are no close counterparts to conditional or causal clauses in the form
of (nonanaphoric) adverbs or prepositional phrases.
von Heusinger, Maienborn and Portner (eds.) 2011, Semantics (HSK 33.2), de Gruyter, 1420–1441
View publication stats

Maienborn Schafer 2011 Adverbsandadverbials

Uploaded by

Copyright:

Available Formats

Maienborn Schafer 2011 Adverbsandadverbials

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Maienborn Schafer 2011 Adverbsandadverbials

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

Adverbs and adverbials

Chapter · November 2011

Claudia Maienborn Martin Schäfer

SEE PROFILE SEE PROFILE

understanding compounds View project

semantics of adverbials View project

The user has requested enhancement of the downloaded file.

Schwarzschild, Roger 2005. Measure phrases as modifiers of adjectives. Recherches Linguistiques

Sigrid Beck, Tübingen (Germany)

54. Adverbs and adverbials

1. Introduction: Towards a deﬁnition of adverbs and adverbials

(1) a. Paul laughed the whole day.

(2) a. Norah treated James *(badly).

(3) a. Norah dresses #(stylishly).

(4) a. He drives extremely/too/very fast.

(5) Joe is provocatively/disappointingly/grotesquely/remarkably stupid.

(6) a. Yesterday was a beautiful day. [Adv as subject]

(8) a. *the well runner

(9) a. now, tomorrow, afterwards [temporal adverbs]

2. Semantic classiﬁcation of adverbials

quantificational and discourse-related adverbials. Before embarking on a more detailed

(10) a. Necessarily, Sam Peckinpah is Sam Peckinpah.

2.1. Predicational adverbials

(11) a. Rudely, Claire greeted the queen.

2.1.1. Sentence adverbials

(12) Peter arrogantly/idiotically put his love letters on the net.

In (12), the speaker judges Peter to be arrogant/idiotic, basing his judgement on

(14) a. Peter arrogantly did not answer my phone call.

Finally, subject-oriented adverbials appear to be anomalous in questions, cf. (15).

(15) ?Did Peter arrogantly not answer my phone call?

(16) a. Honestly/frankly, I have no idea what you’re talking about.

In declaratives, speech-act adverbials allow the addition of the participle speaking

(17) a. Maybe/probably/surely Mary is still alive.

Epistemic adverbials are often referred to as “epistemic modals”, contrasting with

(18) a. *Matthew is not probably dead.

(19) a. Necessarily, nine is an odd number.

(20) a. Necessarily, nine is an odd number.

(21) Fortunately/surprisingly, Peter is back in Australia.

(22) Fortunately/unfortunately, Peter is back in Australia.

(23) It is fortunate/unfortunate that Peter is back in Australia.

(24) a. Peter is fortunately back in Australia.

Evaluatives are also anomalous in questions, cf. (26).

(26) *Is Peter fortunately back in Australia?

(27) a. Emotionally Zardock is cold as ice.

(28) Deixis-wise, this sentence is intriguing.

(28') Deixis-wise, this sentence is intriguing.

2.1.2. Verb-related adverbials

(29) Claire reluctantly/gladly went to school.

(30) Martha gladly did not go to school.

(31) Oedipus intentionally married Jocasta.

(32) Klogman defended himself skillfully/intelligently/hectically.

There is a straightforward semantic explanation for this behavior: Sentence negation

(34) Klogman skillfully didn’t answer the question.

On a manner reading of skillfully in (34), what is skillful is some activity of Klogman

(35) Lochnan loves her very much/deeply.

(37) a. United Stated destroyed Switzerland economically.

(38) The door is halfway closed. ↛ The door is closed.