A STUDY ON EVIDENTIALS IN ENGLISH

PART I: INTRODUCTION
* * *
1. MOTIVATION OF THE STUDY
Natural languages, true enough, offer speakers many and various linguistic devices to
facilitate their communication. That is, these devices are supposed to support the speakers in
terms of sharing information together with expressing their emotions and attitudes. More
importantly, these linguistic devices do give some certain influence over the listeners or the
information recipients’ beliefs or behaviors. These devices fall into the category of
Evidentials – one kind of Epistemic modality.
(1) It sounds like it’s raining. (Evidentials)
(2) The rumor is that she was killed. (Evidentials)
Linguistically, Evidentials are of prime importance in both spoken and written language.
Evidentials, admittedly, are said to come to the speakers’ assistance in expressing well their
certainties, their doubts, their guesses, and their hypotheses in conversations and writings
based upon the certain and absorbed ground of information. In other words, their utterance is
normally said to consist of their attitudes towards the accepted fact in terms of believability,
reliability, and compatibility.
However, not many linguists have formed a distinctively profound study on Evidential
modal markers. Most of the celebrated linguists have paid great attention to discussing
Modality in general and Epistemic modality in particular. Palmer (Mood and Modality,
1986), for example, investigates and restricts his study of Epistemic modality to what is
systematized and organized within the grammatical systems of languages. Whereas, Holmes
(Mood and Modality,1986) presents the expression of Epistemic modality to which is
attached the use of the full range of lexical devices in a variety of written and spoken texts.
Lyons (Semantics, 1977) then offers theoretically possible examples of objective Epistemic
modality together with subjective modality including modal adverbs such as “certainly” and
1
“possibly” mentioned as lexical devices. Givón (Mind, Code and Context – Essays in
Pragmatics, 1989) also shows his interest in modality in a way of producing a theory of
Epistemic scale, meanwhile Halliday (An Introduction to Functional Grammar, 1985)
applies his Theme-Rheme structure to the describing the syntactic functions of Epistemic

markers in a clause as message. Also, among the Vietnamese linguists who prove absorbed
in studying Epistemic modality, Do Huu Chau stands out as a linguist who discusses the
concept of Epistemic modality in the view of pragmatics under his account (Systematic
Semantics – Active Semantics, 1983).
Besides, as far as learners of English are concerned, to master successfully Evidentials is not
an easy task, even for those who are at more advanced proficiency levels. It is well observed
that English learners just focus on the use of some certain Evidentials such as “think, sure,
believe”, which sound popular and are ready on the tip of their tongue. To put another way,
they lack varieties of Evidentials to encode the ground of information in their utterance.
Consequently, that is thought to cause a barrier to the communication co-operation, even the
doubt about the reliability of the utterance. Moreover, the poor use of Evidentials this way
limits them to boring conversations.
Given all the reasons, such a good and informatively full-of-knowledge study on Evidentials
in English is necessary. Thus, the choice of “A study on Evidential Modal Markers in
English” as the subject of the thesis is not accidental.
2. AIMS OF THE STUDY
The study of Evidentials in English is centrally concerned with the following focuses:
i. How evidentiality is expressed by Evidential modal adverbs and adjectives in English.
ii. How evidentiality is expressed by Evidential modal nouns in English.
iii. How evidentiality is expressed by Evidential modal lexical verbs in English.
To achieve the aim, the study will examine three factors – semantic, syntactic, and
pragmatic – that are said to have effects on the use of the expression forms of Evidentials.
2
The study is expected to clarify the relationship between these expression forms and the
difference in the way they express evidentiality.
3. SCOPE OF THE STUDY
Due to the limitation of time, it seems too ambitious to cover all the means to encode
evidentiality in English. Therefore, it is much better and more practical that the study just
centers on pure Evidentials. Hence, a relatively small set of high-frequency Evidential
lexical items which are restricted to our attention appear to stand out as follows:

• Evidential modal adverbs and adjectives: seemingly, apparently - apparent,
evidently - evident, obviously - obviously, surely – sure, undoubtedly, doubtful.
• Evidential modal nouns: rumor, doubt, truth.
• Evidential modal lexical verbs: think, believe, guess, suppose, doubt, see, hear,
taste, feel, smell, appear, seem, say, tell, sound, look.
Despite the fact that the paralinguistic factors such as hesitations, facial expressions, body
gestures, eye movements, etc. play an important role in expressing evidentiality, we find it
impossible to figure them out in this thesis due to the limitation of time. That is the reason
why we study Evidentials in only three aspects: semantic, syntactic, and pragmatic.
In terms of semantic aspect, we will have a focused investigation into the lexical meaning of
the Evidential modal markers. On these grounds, we will put them in order of certainty level,
which proves useful for our study analysis.
As far as syntactic aspect is concerned, we will have a close look at the way the utterances
including Evidential modal markers are grammatically structured. Moreover, the position of
these Evidential modal markers embedded within the utterance grasps our great attention.
3
From the pragmatic aspect, we find it necessary to deal with the conditions that govern the
use of these Evidential modal markers in the process of communication such as speech acts.
Additionally, in order to have a more comprehensive account on the culture-specific aspect
of Evidential modal markers, we will take account of theory of politeness.
4. METHODS OF THE STUDY
Data collection procedure:
With respect to the data presented in the thesis, they include primarily examples collected
from authentic sources such as TV News Programmes at the website of BBC News
() (the programs broadcast on 4
th
- 30
th
April, 2008), and the newspaper
International Herald Tribune, The Global Edition of The New York Times, Issues:

September 5
th
- 14
th
, 2003. These written materials, and T.V News Programmes are all of
common topics found in everyday life. All the data were noted down when we were
watching the TV News Programmes and reading the issues of International Herald Tribune.
The data are collected randomly from these two sources. Yet, the data presented in this study
represent only a fraction of the data considered in developing the proposed analysis. With a
view to serving the study well, utterances used as examples are in declarative form or the
form for statements. A chosen utterance is required to:
• correspond to the expression of an Evidential modal function,
• and involve explicit one or more Evidential modal markers which have been
mentioned as Evidential modal adverbs and adjectives, Evidential modal nouns, and
Evidential modal lexical verbs.
Data analysis procedure:
The theoretical background is based on the theoretical frameworks by different linguists.
Von Wright (1951), Steele (1975), Lyons (1977, 1995), Givãn (1982, 1989), Palmer (1986,
2000), Keifer (1987), etc. propose such well known and convincing researches on which we
4
will rely for the theory of Modality in general, and Epistemic modality in particular.
Meanwhile, the linguists such as Belbert, (1977), Barnes (1984), Anderson (1986), Chafe
(1986), Willett (1988), Bybee (1995), de Haan (1998, 2001), Nuyts (2000), De Lancey
(2001), etc. stand out with multi-dimensional reseaches into Evidentials. That seems to open
a world of references relevant in support of this thesis. The presentation of these linguists’
theory is to give the readers a big picture of Modality, Epistemic modality, and Evidentials.
However, for the main aims of studying and analyzing the thesis semantically, we are going
to take the frameworks by Givãn (1982, 1989), and Palmer (1986, 2000) into consideration
as principal ground of theory on which the Evidential modal markers are analyzed. The
reason is that we look at Evidentials as devices of modality, which is well supported by

Palmer’s theory. Besides, we tend to rank these Evidential modal markers at the scale of
certainty level, which is well proposed by Givãn.
Furthermore, in terms of syntactic aspect, we are going to follow the theory by Quirk (1972),
and Halliday (1985) which forms the basis for the analysis of Evidentials. The two linguists’
frameworks are at my disposal when investigating the position of the Evidential modal
markers located within the utterance and examining the grammatical structures built for the
expressions of evidentiality.
As stated in the Scope of the study, we are going to consider Evidential modal markers in the
context of the process of communication; hence, we will take account of the strategies the
speaker uses when uttering with Evidential modal markers in light of Searle’s (1969, 1985),
and Austin’s (1962) theory about Speech Acts and Brown’s and Levinson’s (1987) theory
about politeness. That may well facilitate our discussion about the pragmatic aspects of
Evidentials in the thesis.
5. DESIGN OF THE STUDY
It sounds appropriate to divide the paper into three main parts:
Part I: Introduction
5
The Introduction presents the background of the study, states what the study is aimed at and
what specific tasks it deals with, identifies the delimitation of the study, and gives a sketch of
methods utilized together with the organization of the study.
Part II: Development
The Development includes 4 chapters:
• Chapter 1 is concerned with the theoretical concepts of Modality, Epistemic
modality, and Evidentials.
• Chapter 2 discusses the semantic features of English Evidential modal markers with
Vietnamese equivalents.
• Chapter 3 presents the syntactic features of English Evidential modal markers.
• Chapter 4 deals with the pragmatic features of English Evidential modal markers.
Part 3: Conclusion
The Conclusion offers an overview or a summary of the study in relation to modality,

epistemic modality in general, and Evidential modal markers in particular. Some
implications relevant are brought forward for learners of English and for further studies.
6
PART II: DEVELOPMENT
* * *
CHAPTER 1: THEORETICAL BACKGROUND
1.1. DEFINITION OF MODALITY
A brief glance at two well-known languages, Latin and English is sufficient to suggest that
the first assumption can be justified. Latin has its systems of mood: indicative, subjective
and imperative; while English has a system of modal verbs: may, can, will, must, etc.
The modal system of most familiar languages is formally associated, along with tense, aspect
and voice, with the verbal systems of the language (and even gender, number and person are
marked on the verb). Yet, modality, as will be seen, does not relate semantically to the verb
alone or primarily, but to the whole sentence. Not surprisingly, therefore, there are languages
in which modality is marked elsewhere rather than on the verb or within a verbal complex.
It is common knowledge that the notion of modality is much vaguer and leaves open a
number of possible definitions. Here is the presentation of some promising definitions by
some celebrated and granted linguists.
1.1.1. DEFINITION OF MODALITY
In Palmer’s theory (Mood and Modality, 1986), modality is defined as semantic information
associated with the speaker’s attitude or opinion about what is said. Whereas, Bybee
(Morphology: A study of the Relation between Meaning and Form, 1985) offers a broader
definition that modality is what the speaker is doing with the whole proposition. Though
these definitions diverge on the particulars, they agree that modality concerns entire
7
statements, not just events or entities, and its domain is the whole expression at the truth-
functional level.
The notion content of modality highlights its association with entire statements; modality
concerns the factual status of information: it signals the relation actuality, validity or
believability of the content of an expression. Modality affects the overall assertability of an

expression and thus takes the entire proposition within its scope. As such, modality evokes
not only objective measures of factual status but also subjective attitudes and orientation
toward the content of an expression by its utterers.
Halliday (An Introduction to Functional Grammar, 1985) views that modality represents the
speaker’s angle, either on the validity of the assertion, or on the rights and wrongs of the
proposal. It is obviously seen that his definition of modality does not diverge much from
Palmer’s and Bybee’s.
For the good sake of a manageable study, it is recommended that Modality herein be
understood in its narrow sense as a semantic term concerning the speaker’s attitude or
opinion to the utterance.
1.1.2. PROPOSITION AND MODALITY
Jespersen (The philosophy of grammar, 1924) talks about the “content of the sentence” and
Lyons (Semantics, 1977) about “the proposition that the sentence expresses”, both wishing
to distinguish them from the speaker’s attitude or opinion. This assumes that a distinction
can be made in a sentence between the modal and the propositional elements, between
modality and proposition.
The distinction between proposition and modality is very close to that of locutionary act and
illocutionary act as proposed by Austin (How to do things with words, 1962). In the
locutionary act we are “saying something”, while in the illocutionary act we are “doing
something” – answering a question, announcing a verdict, giving a warning or making a
promise. These ideas are on the basis of speech act theory.
8
Lewis (An analysis of knowledge and evaluation, 1946) proposes that “the proposition is
assertable; the contents of the assertion…can be questioned, denied or merely supposed,
and can be entertained in other moods as well.” But “modality” in this sense, referring to all
the non-propositional elements of a sentence, is much wider than in the sense in which it will
be used here.
Similarly, Rescher (Topics in philosophical logic, 1968) talks about propositions and argues
that where a proposition (which may be true or false) is subject to further qualification, this
qualification represents modality.

1.1.3. TYPES OF MODALITY
The distinction that Jespersen (The philosophy of grammar, 1924) draws between his two
sets: “containing an element of will and not containing an element of will” is closely
paralleled in Lyons’ reference to “the speaker’s opinion or attitude towards the proposition
described”. Lyons, in conjunction with other scholars, recognizes two kinds of modality,
using Wright’s terms:
Epistemic modality, which is concerned with matters of knowledge, belief or opinion rather
than fact.
Deontic modality, which is concerned with the necessity or possibility of acts performed by
morally responsible agents.
Steele et al. (An encyclopedia of AUX: a study in cross-linguistic equivalence, 1981)
implicitly make the same distinction: “Elements expressing modality will mark any of the
following: possibility or the related notion of permission, probability or the related notion of
obligation, certainty or the related notion of recruitment.” The remarks in Steele et al. can
be illustrated from English; the following sentences can be interpreted either in terms of
possibility, probability, and necessity, or in terms of permission, obligation and requirement:
(3) He may come tomorrow.
9
(Perhaps he will/ He is permitted.)
(4) The book should be on the shelf.
(It probably is/ Its proper place is.)
(5) He must be in his office.
(I am certain that he is/ He is obliged to be.)
Lyons (1977) also suggests a distinction between objective modality and subjective modality.
He presents a typical example:
(6) Alfred may be unmarried.
It can be interpreted that the speaker may be understood as subjectively qualifying his
commitment to the possibility of Alfred’s being unmarried in his own certainty terms and the
sentence is more or less equivalent to “perhaps Alfred is unmarried”. There are, however,
situations in which the possibility of Alfred’s being unmarried is presentable as an objective

fact. The speaker might reasonably say that he knows, and does not merely think or believe,
that there is a possibility of Alfred’s being unmarried. In his words:
“Objective modality refers to reality; it is part of the description of the world. Subjective
modality, on the other hand, is the expression of the speaker’s beliefs.”
These are two kinds of epistemic modality. The semantic differences between subjective and
objective epistemic modality are significant. Objective modalized sentences are statements of
fact; thus, they can be denied and questioned while subjective epistemic sentences, express
the speaker’s beliefs and not statement of fact; hence, they cannot be denied or questioned.
1.2. EPISTEMIC MODALITY
1.2.1. DEFINITIONS OF EPISTEMIC MODALITY
10
The term “epistemic” is suggested to be applied to not only modal systems that basically
involve the notions of possibility and necessity, but to any modal system that indicates the
degree of commitment by the speaker to what he says. In particular, it should include
Evidentials such as “hearsay” or “report” (the Quotative) or the evidence of the senses. The
Declarative, moreover, can be regarded as the unmarked (“unmodalized”) member of an
epistemic system, though by this definition some languages have no “unmodalized”
declaratives.
This use of the term may be wider than usual, but it seems completely justified
etymologically since it is derived from Greek word meaning “understanding” or
“knowledge” (rather than “belief”), and so is to be interpreted as showing the status of the
speaker’s understanding or knowledge; this clearly includes both his own judgments and the
kind of warrant he has for what he says.
It is often claimed in the linguistics literature that epistemic modality, unlike other kinds of
modality, does not contribute to the truth conditions of the utterance. Relatedly, several
commentators argue that epistemic modality expresses a comment on the proposition
expressed by the rest of the utterance:
Epistemic modality… is the speaker’s assessment of probability and predictability. It is
external to the content, being a part of the attitude taken up by the speaker: his attitude, in
this case, towards his own speech role as ‘declarer’. (Halliday, 1970: 349)

Epistemic modality indicates … the status of the proposition in terms of the speaker’s
commitment to it. (Palmer, 1986: 54-5)
Epistemics are clausal-scope indicators of a speaker’s commitment to the truth of a
proposition. (Bybee & Fleischman, 1995: 6)
Epistemic modals must be analyzed as evidential markers. As such they are part of the
extra-propositional layer of clause structure and take scope over all propositional operators...
(Drubig, 2001: 44)
11
The intuition underlying this view is that epistemic modality in natural language marks the
degree and/or source of the speaker’s commitment to the embedded proposition. According
to this view, the proposition expressed by the utterance in (9) can be paraphrased by (10); the
modal force of the utterance indicates that the speaker entertains the embedded proposition
with a low degree of commitment:
(7) John may be at home.
(8) John is at home.
This position on epistemic modality is at odds with several semantic treatments of modality,
in which epistemic modality (alongside other types of modality) is seen as regularly
contributing to truth conditions. On these accounts, modal operators in natural language
encode modal force (necessity or possibility) which gets relativized with respect to different
types of contextual assumptions or conversational backgrounds (Kratzer, 1981; 1991; cf.
Lewis, 1986; Brennan, 1993; Papafragou, 2000). Depending on the specific conversational
background selected, modal expressions receive different kinds of interpretation, as shown
by the paraphrases in the examples below:
(9) The children must be leaving.
(10) a. In view of what the speaker knows, the children must be leaving.
b. In view of what their obligations are, the children must be leaving.
(11) John may leave.
(12) a. In view of what the speaker knows, John may leave.
b. In view of what the circumstances are, John may leave.
The interpretations in (9a), (12a) involve epistemic conversational backgrounds, while those

in (9b) or (12b) involve deontic (or root) conversational backgrounds. In this kind of theory,
deontic and epistemic modality are treated symmetrically, and both are seen as contributing
to the proposition expressed by the utterance.
1.2.2. TYPES OF EPISTEMIC MODALITY
12
It is well known that Palmer claims epistemic modality is divided into two basic categories:
Judgements and Evidentials. It is can be vividly described in the diagram below:
Epistemic modality
Evidentials Judgments
Type a Type b Inference Confidence
“certainty” -
Knowledge
Visual [Declarative] [Declarative] [Declarative]
– Belief
Non-visual Sensation
Report
Deductive Deductive
Assumptive Assumptive “probably”
Spectaculative “possibly”
Diagram 1: Types of Epistemic modality
(Palmer, 1986 : 95)
Notes:
[ ]


unmarked member
equivalence
“stronger than”
As shown in the diagram, Judgments and Evidentials are stated to be concerned with the
indication by the speaker of his commitment to the truth of the proposition being expressed.

Concretely, the former includes all epistemic notions that involve possibility and necessity,
particularly with regard to speculation and deduction on the part of the speaker as subject or
perceiver of the information. Judgments assert the possibility of the truth of a proposition
without any overt indication of the grounds for the assertion. Judgments can be categorized
13
by the degree of confidence that the speaker has in the assertion, which produce two
subcategories: necessary judgments and possible judgments, respectively based upon
inference and confidence, deduction and speculation, or strong and weak judgment. It is
observed that Judgments in English, which have a variety of modal auxiliaries to indicate the
speaker’s assessment of the content of the proposition. It is well described in these following
examples:
(13) Michael might lead the league in scoring next season.
(14) There must have been many chairs in that room.
In contrast to Judgments are Evidentials, which encode the grounds on which a speaker
makes an overtly qualified assertion. Unlike Judgments, Evidentials explicitly signal the
collateral that a speaker takes as substantiating an assertion. The concepts as well as the
types of Evidentials will be discussed further and detailedly presented in the next part.
1.3. CONCEPTS OF EVIDENTIALS
1.3.1. DEFINITIONS OF EVIDENTIALS
As mentioned above, Evidentials encode the grounds on which a speaker makes an overtly
qualified assertion. To put it in another way, Evidentials encode the speakers’ source of
information as well as indicate the reliability of the information. They put in perspective or
evaluate the truth-value of a sentence both with respect to the source of the information
contained in the sentence, and concerning the degree to which this truth can be verified or
justified. Synonymously, Evidentials illustrate the type of justification for a claim that is
available to the person making that claim. This justification can be expressed by markers
referring to immediate evidence on the basis of visual observation, to inference on the basis
of (non)observable facts, to deduction or inference, etc. In “Chafe & Nicholas” (1986),
briefly, they represent a “natural epistemology”.
14

More clearly stated, Evidentials also are to relativize or measure the information status of the
sentence. The term “information status” is intended to include both the truth-value of a
sentence and the relative importance accorded to it. Evidentials are said to own itself two
essential properties. First and most importantly, a source of evaluation or reliability of the
sentence is involved. This “source of information” defines who stands for the information
status of the sentence. Secondly, the information status of the sentence is most often
measured on the type varies: the sentence is measured with respect to reliability, probability,
expectation or desirability.
1.3.2. DEFINITIONS OF EVIDENTIAL MODAL MARKERS
Evidential modal markers are defined as lexical categories which indicate how and to what
extent speakers stand for the truth of the statements they make. In English, all Evidential
modal markers are lexical. English does not have grammaticalized Evidentials, but there are
lexicalized Evidentials. The lexicalized Evidentials of English include the Evidential modal
senses of verbs of expression and verbs of appearance, and Evidential adverbs such as
“evidently and apparently”. Other language units for Evidential modal markers are about to
be well-mentioned later.
1.3.3. TYPES OF EVIDENTIALS
In the views of different linguists, Evidentials can be divided in different ways. Jakoson
(1957) represents four types of Evidentials in terms of the source of information:
• quotative (hearsay),
• revelative (dream),
• a guess (presumptive), and
• previous experience (memory).
For Carib, Hoff (1991) distinguishes introspective Evidentials (knowledge from inference)
from extraspective Evidentials (culturally available knowledge).
15
Taking Evidentials into consideration, Palmer (1986) observes that there are at least four
ways of presenting a statement or a fact:
• speculative,
• deductive,

• hearsay,
• appearance based on sensory evidence.
Barnes (1984) suggests that visual, nonvisual (sensory), apparent, secondhand, and
assumed be types of Evidentials.
Generally known, Evidentials can encode speaker-oriented qualifications of propositions
along two dimensions:
• in terms of the evidence they are based on
• with respect to the speaker’s commitment to their truth
The two dimensions – whether the proposition is based on perceptual evidence and whether
the speaker believes in its truth – are logically independent. Yet natural language typically
treats propositions based on perceptual evidence on the part of the speaker as propositions
asserted by the speaker to be true. Similarly, if a speaker marks a proposition as based not on
first-hand evidence, he/she typically expresses non-commitment to the truth of the
proposition. That is why it is more useful to think of the terms Direct and Indirect – the two
subcategories of Evidentials – as making a distinction based not on whether or not the
evidence is perceptual, but on whether or not the evidence justifies the speaker’s belief in
proposition.
16
Evidentials
Indirect
Evidentials
Direct
Evidentials
includes all the markers of the
speaker’s firsthand evidence such as
visual, auditory, and other sensory
modalities, with visual evidence by far
the most reliable.
encompasses all forms of secondhand
fact such as report, quotation, hearsay,

assumption, appearance, and all other
types of supportive, auxiliary
information, of which quotation and
hearsay are found the most common.
Diagram 2: Types of Evidentials
(15) I heard him beating the dog at 7 p.m. yesterday.
(Direct Evidentials)
(16) Rumor has it that the police have arrested John.
(Indirect Evidentials)
1.3.4. SCALES OF EVIDENTIALS
No doubt arises that scales, and more generally hierarchies, have proven useful tools in
typology as well in semantics and pragmatics. In typological research, the most common
type of hierarchy are implicational universals which are used to predict possible and
impossible language systems as well as directions of diachronic language change, among
other things (Croft ,1990, and Dik, 1981). In semantics and pragmatics, linguistics scales are
taken to be an ordering of a set of linguistic expressions belonging to a single grammatical
category, where the order is determined by degree of informativeness or semantics strength
(Levinson, 2000). These scales are generally used to explain conversational implicatures
(Grice, 1989) associated with the linguistic expressions they order.
17
In terms of Evidential scales, Willet (1988) proposes the hierarchy as follows:
ATTESTED > REPORTED > INFERRING
In this hierarchy, he explains that “on the scale from most to least direct, Attested evidence is
ranked as the most reliable source, Inferring evidence as the least reliable, and Reported
evidence somewhere in the middle”. (Willett, 1988:86)
Thus, for Willett the speaker’s preference of certain evidence types over others is based on
two criteria: directness and reliability. He goes on to say that “a speaker using on Inferring
evidential denies having reported or direct evidence,” that is, for him the indirect negation
between Inference and Reportative goes in the opposite direction.
Meanwhile, it is seen that de Hann (1998) shows another scale for Evidentials:

VISUAL > NONVISUAL > INFERENTIAL > QUOTATIVE
Here, de Hann orders Inferential above Quotative. He also uses two ordering criteria, the
first of which is also directness. However, his second criterion is not reliability. With respect
to the relative ordering of Inference and Quotative, de Hann (1998) states:
“Within the area of indirect evidence, Inference is closer to direct evidence than Hearsay
because by using a Quotative, the speaker relies wholly on evidence that comes from
another source. The Inferential is used when the speaker is involved himself or herself with
the evidence to a certain degree. The speaker makes deductions on the basis of evidence.
This evidence has been collected by the speaker, which makes him or her more of and active
partner than in the passive act of receiving information from another source.”
Thus, de Hann’s second criterion is speaker involvement. Willett and de Hann, therefore,
agree on the relative orderings derived from directness, namely that all kinds of direct
18
evidence should be ordered above all kinds of indirect evidence, but their different second
criteria lead them to postulate different relative orderings of Inference and Reportative.
To contribute to building up the scales of Evidentials’ certainty, Givãn argues that languages
qualify evidence along four gradients. It is clearly shown in the diagram below:



Diagram 3: Gradients of quantified evidence (Givãn, 1989)
Moreover, Givãn argues that there are three kinds of propositions (P), typed by their inherent
certainty and need for substantiation:


19
The High Certainty
The Medium Certainty
The Low Certainty
P are taken for

granted,
presupposed and
above challenge.
Here, the
presupposed
information are
assumed by the
speaker to be
known to, familiar
to, or otherwise
unlikely to be
challenged by the
hearer.
P are doubtful
hypotheses and
beneath challenge
and substantiation.
P here are known
as irrealis. The
information whose
source is largely
irrelevant is
weakly asserted.
The speaker does
not intend to
defend the
information in
order to solicit
challenge,
correction or

corroboration.
P are open to
challenge and thus
require supporting
evidence. P are
known as realis-
assertion. Under
this mode,
information is
strongly asserted,
yet it remains open
to challenge by the
hearer. The speaker
must then be
prepared to defend
the information by
citing the source of
evidence.
Evidence shown by
Person
Sense
Directness
Proximity
Speaker
Vision
Senses
Near Far
Inference
Hearing
Other

senses
Feeling
3
rd
person
Hearer
Diagram 4: Scale of Evidentials’ certainty (Givãn, 1989)

If a speaker is forced to choose evidence to defend his assertion, he chooses evidence
according to the four scales presented in the diagram, and according to the internal order of
the gradients, vision over hearing, for instance.
Givãn provides rules of evidence for his scale and points out that only in the case of realis-
assertion is evidence assumed to be both available and expected, which is ranked according
to the degree of evidentiary strength. It is also claimed that in languages further
differentiating among several sensory sources of direct evidence, the grammar of
evidentiality tends to rank the senses according to their reliability as source of evidence.
Besides, in the grammar of evidentiality, one finds the ranking of either the participants in
the event according to person or the temporal proximity of the reported event to the speech
time in a rather predictable way. The diagram below does help to illustrate the scales:
20
Scales
Scale of
temporal
proximity
Scale of spatial
proximity
Scale of
participants in
events
Scale of

reliability of
sensory
evidence
Scale of
evidentiary of
source
a. Direct
sensory
experience
b. Inference
from direct
sensory
evidence
c. Indirect
inference
d. Hearsay
a. Visual
experience
b. Auditory
experience
c. Other
sensory
experience
a. Speaker
b. Hearer
c. Third
party
a. Near
speech
situation

b. Away
from speech
situation
a. Nearer to
speech time
b. Farther
away from
speech time
Diagram 5: Types of scales (Givãn, 1989)
1.3.5. EVIDENTIALS AND INTERACTION
Various different types of interaction may arise between Evidentials and different values of
person. In particular, interactions may affect the frequency or the interpretation of
Evidentials, or both. These interactions can be considered to fall into four types:
• (1) The frequency of certain Evidentials varies greatly depending on person.
For example, reported Evidentials are uncommon with first person because
normally when a speaker was involved in an action, they do not need to be
told that it occurred.
• (2) The interpretation of some Evidentials in some languages changes
depending on the person values in the sentence. For example, with a first
person subject, non-witnessed Evidentials often encode that an action was
21
non-volitional, rather than indicating that the speaker did not witness the
event, their more common interpretation.
• (3) Particular Evidentials and person values cannot co-occur in particular
languages. For example, in some languages an inferred evidential cannot be
used in first person contexts.
• (4) Certain Evidentials can be used with utterances which involve any person,
but with restrictions on particular combinations of person and evidential with
particular predicate types. For example, a predicate indicating something
about an internal state may be used with a non-visual evidential, showing a

speaker’s knowledge is based on feeling - but only if the internal state relates
to the speaker, since otherwise the speaker cannot know about the state
through having felt it.
These four types of interaction are not fully independent — for example, it is precisely when
a particular combination of person and evidential is infrequent that the evidential is likely to
develop a distinct interpretation. However they are a useful schema for examining and
classifying types of interaction.
1.4. THEME - RHEME STRUCTURE AND EVIDENTIALS
Halliday (An Introduction to Functional Grammar, 1985) observes that in all languages, the
clause has the character of a message: it has some form of organization giving it a status of a
communicative event. In English, as in many other languages, the clause is organized as a
message by having a special status assigned to one part of it. One element in the clause is
enunciated as the Theme; this then combines with the remainder so that the two parts
together constitute a message.
In Halliday’s reference, the Theme is the element which serves as the point of departure of
the message; it is that with which the clause is concerned. The remainder of the message, the
22
part in which the Theme is developed, is called the Rheme. As a message structure,
therefore, a clause consists of a Theme accompanied by a Rheme; and the structure is
expressed by the order – whatever is chosen as the Theme is put first.
As a general guide, the Theme can be identified as the element which comes in that first
position in the clause to be the starting point for the message; it is what the clause is going to
be about. Additionally, the Theme can be nominal group, an adverbial group, or a
propositional phrase.
Theme Rheme
(17) The duke has given my aunt that teapot
(18) Very carefully she put him back on his feet again
(19) With sobs and tears he sorted out those of the largest size
As for the Modal markers in general, and Evidential modal markers in particular, Halliday
considers them the Modal adjuncts which are pointed out to be thematic, but not to be

obligatory so. Besides, within the structure Evidential modal constructions can be found in
many positions. Their mobility is well seen in the following examples:

Initial Medial Final
(20) I am sure (that) Peter (Theme) can pass the exam. (Rheme)
(21) Peter, (Theme) I am sure, can pass the exam. (Rheme)
(22) Peter (Theme) can pass the exam, (Rheme) I am sure.
1.5. BACKGROUND ON SPEECH ACT THEORY
1.5.1. SPEECH ACT THEORY
This section offers a brief introduction to the basic speech act theoretic concepts developed
by Searle and his colleagues (Searle and Vanderveken, 1985, Vanderveken, 1990,
Vanderveken, 1991.) who based them on ideas first introduced by Austin (1962).
Austin (1962) presents three acts the speaker performs when he/she utters a sentence:
23
• Locutionary act
By locutionary act, the speaker selects language units as phonetic units, lexical
items, grammatical rules and combines these to form an utterance.
• Illocutionary act
By illocutionary act, the speaker is using a sentence to perform a function. He may
use some specific language units as indicators to signal the function of the utterance.
• Perlocutionary act
By perlocutionary act, the speaker conveys an idea, a further purpose which can be
interpreted by the hearer, not on the surface of the words and structures of the
sentence by means of his manner of locutionary act in an actual situation.
Here come the basic assumptions of speech act theory which are summarized in the
following quote from Searle and Vanderveken (1985).
The minimal units of human communication are speech acts of a type
called illocutionary acts (terminology introduced by Austin (1962)).
Some examples for illocutionary acts are statements, questions,
commands, promises, and apologies. Whenever as speaker utters a

sentence in an appropriate context with certain intentions, he performs
one or more illocutionary acts. In general an illocutionary act consists of
an illocutionary force F and a propositional content P. For example, the
two utterances “You will leave the room” and “Leave the room!” have
the same propositional content, namely that you will leave the room; but
characteristically the first of these has the illocutionary force of a
prediction and the second has the illocutionary force of an order. (Searle
and Vanderveken, 1985:1)
An illocutionary force F is a complex entity. According to Vanderveken (1990:103), it has
the following six components:
• an illocutionary point,
24
• a mode of achievement of an illocutionary point,
• propositional content,
• preparatory and sincerity conditions,
• and a degree of strength.
The illocutionary point indicates how the propositional content of the illocutionary act
relates to the world. For assertions, the illocutionary point is to match the propositional
content to the world; for other speech acts such as directives, it is to match the world to the
propositional content. Vanderveken (1990: 105) recognizes five illocutionary points,
following Searle and Vanderveken (1985):
• the assertive point which consists of representing as actual a state of affairs;
• the commissive point which consists of committing the speaker to a future course of
action;
• the directive point which consists of making an attempt to get the hearer to do
something;
• the declarative point which consists of performing an action which brings into
existence a state of affairs by representing oneself as performing that action;
• and the expressive point which consists of expressing propositional attitudes of the
speaker about a state of affairs.

Illocutionary forces have three types of pre-conditions as components:
• Propositional content conditions put restrictions on the propositional content of an
illocutionary act of a particular force. For example, the propositional content of a
promise must present a speaker’s future course of action.
• Preparatory conditions are certain propositions that speaker takes for granted in
performing the illocutionary act.
• Sincerity conditions are “propositional attitudes of the form m(P), where m is a
psychological mode such as, for instance, desire, regret, or hope […] A performance
of an illocutionary act is sincere when the speaker has the mental state that he
25