A DESCRIPTIVE ANALYSIS OF THE USE OF FILLED PAUSES
AMONG NATIVE AND NONNATIVE SPEAKERS OF ENGLISH
JoELLEN M. SIMPSON
DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
JoEllen M. Simpson
I would like to thank my committee members for their support and
coaching through this project.
As chair, Dr. Roger Thompson was able to
steer me in the right direction to make a stronger dissertation.
Boxer, as cochair, was invaluable for her information on methodologies and
especially like to thank Dr. Norman Markel for giving me moral support,
, and help with the statistical analysis.
for sticking with me
I thank Dr. David Pharies
of the changes and also thanks to
Patricia Craddock for being the outside member.
thanks go to my parents,
Marita and Philip Simpson, for
their support through the years.
They have provided the moral and mental
foundation needed for me to make it through all the years of school.
finally thanks go to
Renteria for his undying support through
emotional roller coaster of preparing this dissertation.
Although many people helped make this dissertation a reality, the
mistakes that remain are mine alone.
TABLE OF CONTENTS
1 AN INTRODUCTION TO THE FUNCTIONS
OF FILLED PAUSES 1
Functions of Filled Pauses 7
Providing Planning Time 7
Managing Turn-Taking 13
Analysis of a Speech Sample 14
The Search for Other Functions of Filled Pauses 21
Outline of Subsequent Chapters 23
2 ETHNOGRAPHIC EXPLORATIONS 26
Part One: Defining a Three-Function Analysis of
Filled Pauses 29
Providing planning Time 30
Managing Turn-Taking 32
Softening for Politeness 34
Part Two: Exploring Three Samples 39
Situation One: Discussing Body Odor 39
Situation Two: Asking for a Favor 46
Situation Three: Responding to Personal
Summary of Ethnographic Responses 54
3 FILLED PAUSES AS PLANNING IN NATURALLY
OCCURRING CONVERSATIONS 55
Methods of Analysis 60
Clause-Boundary Planning 62
Managing Turn-Taking 72
Topic Switching 75
Side Sequences and Interruptions 80
Clause-Internal Planning 84
Summary of Filled Pauses as Planning 94
4 FILLED PAUSES AS SOFTENING IN NATURALLY
OCCURRING CONVERSATIONS 95
Responding to Personal Questions 1(B
Miscellaneous Sensitive Topics K(B
Other Repair 118
Conversational Closings 119
Summary of Filled Pauses as Softening 121
5 DISCUSSION AND IMPLICATIONS OF FILLED PAUSES
IN NATIVE AND NONNATIVE SPEECH 123
Numerical Analysis 127
Theoretical Implications 137
Pedagogical Implications 14)
A TRANSCRIPTION KEY 145
B TRANSCRIPTS OF ETHNOGRAPHIC INTERVIEWS 147
C SAMPLE OF TRANSCRIPT FROM ANALYSIS OF
BIOGRAPHICAL SKETCH 179
Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
A DESCRIPTIVE ANALYSIS OF THE USE OF FILLED PAUSES
AMONG NATIVE AND NONNATIVE SPEAKERS OF ENGLISH
JoELLEN M. SIMPSON
Chair: Roger M.
Major Department: Linguistics
This dissertation presents an analysis of the use of filled pauses in the speech
of native speakers of American English and nonnative learners of American English.
A filled pause is defined as a nonlexical sound that occurs as a hesitation in speech.
considered to function only to provide a speaker with thinking time while speaking.
ethnographic interviews of naive native English speakers provide the basic functions
of the filled pause as used by this
group of speakers.
A subsequent analysis of
conversational dyads provides evidence that corroborates and adds to the functions
described in the ethnographic interviews.
Based on the information collected from published literature, plus the two
ft -_ ,-. J 4 .**/
Two main uses of filled pauses emerge:
2) to make a sensitive situation more pol:
1) in planning for forthcoming speech and
ite. Overlapping the planning function we
see planning filled pauses used to manage turn-taking, to switch topics, and to
recover from side sequences or interruptions.
Similarly, the function of softening
covers a variety of situations, from asking and responding to favors, to answering
personal questions, to closing conversations, to repairing other speakers' mistakes, and
to dealing with many other assorted taboo topics, ranging from alcohol to bathrooms.
When the speech of native and nonnative speakers is examined, filled pauses
are found to be much more common in nonnative speech than in native speech.
main difference between the two groups is in the distribution of filled pauses in the
two main functions.
Nonnative speakers use filled pauses mainly for gaining planning
with very little use of filled pauses in sensitive situations.
On the other hand,
native speakers use filled
for both functions.
Implications for the
ethnographic interviews in ESL research are examined, and the importance of filled
pauses in ESL pedagogy is also discussed.
AN INTRODUCTION TO THE FUNCTIONS OF FILLED PAUSES
Every person who has tried to learn a second or foreign language
knows the difficulties of speaking as fluently as a native.
Some try to build
fluency by focusing their attention only on grammatical constructions and
pronunciation of new sounds.
, fluency is simply the ability to
speak rapidly without any hesitations.
But is fluency more than speed and
comprehensibility? How does the speaker take time to plan for forthcoming
without yielding the
floor to the other speakers?
various types provide that thinking time.
Without them, speakers cannot
absence of hesitations could make comprehension more difficult because
there is not enough time to process the incoming information.
hesitations can even pass on subtle meanings to listeners. O
type of hesitation of interest for us is the filled pause.
functions? Basically, the definitions that have been given thus far fall short
of providing us with an understanding of filled pauses and what they are
There have been many interpretations and little agreement even
about the physical manifestation of filled pauses.
For example, Goldman-
Eisler (1961) defined filled pauses as simple sounds of hesitation, i.e
sounds: /a, e,
Faerch and Kasper (1983:215) described filled
pauses as pauses that involved some nonlexical vocal cord activity like
er, erm, uh."
one of the better definitions
, Dollaghan and Campbell
(1992) suggested that filled pauses were "conventional but nonlexical,
filler vocalizations, such as um and er" (61).
important to explain
the significance of hesitations
general and to describe the role of filled pauses in native speech.
and Rehbein (1987), have discussed how native speakers are able
reactions from listeners.
These researchers noted that a great amount of
including the use of filled
explained that as much as one third to one half of speaking time could be
made up of hesitations (including silent and filled
Goldman-Eisler wrote that "evidently pausing is as much a part
of the act of sDeakine as the vocal utterance of words itself,
community and [were] often not even consciously noticed by the hearer"
He noted that they were simply a natural part of the encoding and
Rehbein (1987:102) also noted that native speaker use
and other features, are granted by the
therefore not taken as indications of nonfluency)."
By such means, native
speakers were able to gain time to plan their next utterance without being
As an example of this,
, and Bilous (1991) observed professional speakers using as many
as 3.17 'urns' per minute in a 50-minute lecture, with no negative responses
"listeners are almost entirely insensitive to the use of 'ums'
and that their
impressions of both
the speaker and the message are
unaffected by the
frequency with which filled pauses are used."
Clearly if native speakers
need time to plan and to understand speech, nonnative speakers have at
least the same need and probably more.
Given the above work on filled pauses,
it seems clear that a certain
number of filled pauses are forgiven as a natural part of native speech.
question is, do nonnative speakers who use filled pauses enjoy the same
strategies, including filled pauses, in nonnative speech.
For example, Darot
interviewed considered hesitations in nonnative speech to be unacceptable.
There are people who believe that filled pauses in the speech of nonnative
speakers are signals of poor speaking ability.
However, it seems clear that
native interlocutors and teachers of second languages should not
As with native speech, hesitation markers probably have an
important role in the language use of the learners as well.
Weinreich (1953), and Rehbein (1987) reported that
while attempting to "quantify the
characteristics of fluency" for native and
filled pauses as "dysfluency markers."
On the other hand, he called silent
or unfilled pauses "temporal elements," which provided planning time for
the speaker. By labeling filled pauses dysfluency markers, Lennon reflected
the belief that filled pauses were in some way undesirable.
nonnative speakers the benefit of the doubt when they were speaking.
pointed out that some Americans were "unduly intolerant of some features
features as used by a] native speaker" (394).
hesitation markers such as pauses, repetition
He noted specifically that
Is and self corrections were
"more likely to be interpreted as dysfluency" if produced by a nonnative
speaker than if produced by a native speaker.
It appears that this has been
a typical attitude of native speakers, biased against the nonnative speakers,
resulting in something like a double standard, or a different interpretation
of elements of speech based on whether the speaker is native or nonnative.
Weinreich (1953:21) discussed this same problem of
listener expectations of
speakers were conversing,
they had expectations that their partner would
have native grammar and pronunciation as well as native use of nonobvious
speaker talked to a native speaker, the nonnative speaker frequently did
not fulfill the expectations of
the native speaker.
Problems occurred when
native speaker often
tolerate differences in grammar, pronunciation, or even filled pause use in
nonnative speaker fluency.
, the perception of speech
nonnative speakers rested not only on the actual production of the speech,
researchers were able to find positive responses by native speakers about
the use of hesitations in nonnative speech.
For example, Raupach (1980)
directly from her L1 (French) into her L2 (English),
was not a highly fluent speaker
and even though she
she was perceived by the native speakers
of the target language as being more comprehensible.
D'Anglejan, and Sankoff (1987) studied fluency of English
and French bilinguals,
particularly with regard to discourse markers.
found that if the speakers avoided long silent pauses
pauses or even lexical fillers (such as like,
perceived as being more fluent. Derwing
by inserting filled
you know), the speakers were
(1990) corroborated this claim
with test results that showed that an increase in the number of pauses did
not adversely affect native speaking listeners; however, increased pause
length was disruptive for listeners.
Again, with a focus on the listener, Scanlan (1987) hoped that a more
native-like use of filled pauses would result in increased comprehensibility.
After observing American learners of French,
Scanlan realized that filled
pauses surfaced precisely when they [were] least welcome, at moments
of linguistic distress" (350).
Scanlan knew that native speakers regularly
verbal pauses would greatly increase the native-like French perception of
, he taught American students of French
how to use filled pauses like native French speakers.
He suggested that
eliminating one distracting element (the
expectations (the French nonlexical filled pause) would allow a learner time
to think without distracting or disturbing the native speaking interlocutor.
Functions of Filled Pauses
If appropriately used filled pauses in the speech of both native and
nonnative speakers may be valuable, what does the literature say about the
functions of filled pauses? Are filled pauses purely random, or is there a
systematicity to their use? Researchers have found filled pauses are used
primarily to provide planning time. Some scholars have also noted that
some planning pauses have the added function of manipulating turn-taking.
In these cases
, the primary function appears to be for planning, but if the
filled pause occurs in a place where speaker turns frequently shift (clause
or sentence boundary positions), there is often the added function of saving
Providing Planning Time
Rehbein (1987) proposed several hypotheses about how hesitations
As a result,
"fluency depends on the activity of planning
an entire utterance in advance" (99).
He wrote that "a second language
utterance which allows him
or her to build in
different parts of speech
while proceeding" (99).
As a result of this simultaneous work, the speakers
Rehbein observed had a need to hesitate while planning ahead.
By looking at what has been written about the placement of filled
pauses for planning,
we can establish that with this
kind of superficial
classification or labeling.
Maclay and Osgood (1959) sparked a tradition of looking at where
various pause markers occurred and what they signaled.
naturally occurring speech, the authors reported that silent pauses occurred
more frequently before content words,
both content and function words. Boon
while filled pauses occurred before
ler and Dittmann (1962:216) defined
of pauses used for planning:
pauses (or clause-
which occurred at terminal junctures, and hesitation
pauses (or clause-internal pauses),
which occurred in all other locations.
Later research by Boomer (1965) suggested that juncture pauses did not
always neatly occur at syntactic boundaries, but instead tended to occur
Lalljee and Cook (1969) and Cook, Smith, and Lalljee (1974) reported
that planning complex syntactic patterns influenced the number of filled
Unfortunately, although they reported that they were intuitively
sure that a relationship existed between complexity and increased filled
their research did not strongly corroborate this notion.
(1980), on the other hand, found that there was a link between complexity
and the use of filled pauses in his Pear Story narratives.
He explained that
, indicating more complexity when planning a shift in topic than
with maintaining a topic.
Hawkins (1971) also proposed that the variety
choosing what to say
and how to say it
resulted in more pause time occurring at clause boundaries.
that it reflected the planning that went into speech.
Holmes (1988) obtained the opposite results.
He reported that in one
before embedded clauses than
before combined clauses.
suggested that this might have been a result of pre-planning of sentences
with embedded clauses
, while sentences with multiple independent clauses
were not pre-planned.
According to these researchers,
the presence of hesitations gives us
as "the result of universal human difficulties in organization and memory"
Likewise, Dollaghan and Campbell (1992:56) noted that different
interpreted as evidence that the speaker [was] encountering significant
information-processing [or planning] demands."
As for nonnative speaker hesitations, some researchers have studied
the perception of the speaker's
of the filled pause.
fluency, which varied based on the position
For example, Schmidt (1992) noted that there existed
different types of planning for all second language learners.
non-fluent learner's pauses,
He wrote that
false starts, and other signs of hesitation
reflect the need to focus attention on the lower levels of planning
native speakers in exhibiting hesitation primarily as a reflection of
integration and macroplanning [clause-boundary planning]. (Schmidt
boundaries or in juncture positions, while less advanced speakers hesitated
more frequently and in clause internal positions.
Yorozuya (1982) described
the same phenomena in his Japanese university students of English as a
The contrast in placement between clause-boundary and
is the clue
fluency of nonnative speech.
Less advanced speakers,
because of their use
of filled pauses in clause internal positions, are judged to be less fluent
understanding of how filled pauses used in different ways could influence
the listener's perception of the fluency of the speaker.
In her data, fluent
without appearing to
In more difficult situations (for example,
the first time an idea
was expressed), there was a tendency to use more clause internal pauses,
which caused the speakers to be perceived as being more dysfluent.
a different research
French students in both their first language and English as their second
found that the
number of filled pauses
second language of his subjects,
but only at the ends of sentences or in
other words, in clause-boundary positions.
Dechert (1980) studied hesitation markers as indicators of internal
verbal planning in nonnative speakers.
He looked at the speech of a single
German student of English as a second language before and after a three
month intensive language study.
He found that the speaker's language
planning was better at the end of the intensive course,
as reflected by his
better global planning strategies.
, rather than trying to classify or locate planning
conversational analysis techniques to get a broader picture of filled pauses
as planning markers.
They were interested in the relationship between
various nonverbal cues and the planning of speech.
(1972) examined the use of the interjections "uh" and "oh."
data were not collected from naturally occurring speech but from sentences
that she made up, she offered conclusions about the different functions of
these two interjections.
Her argument rested on sentences such as
2. I saw
- twelve people at the party.
- twelve people at the party.
in which the meaning of each was clearly different (#1 signalled the need
to plan the sentence; #2 signaled an estimation of the number of guests).
functions for these filled pauses.
, Schachter, and Bilous (1991) described an intriguing
correlation between gestures and filled pauses.
Their research showed that
when native speakers were gesturing, they were less likely to use filled
The authors suggested that when speakers used gestures,
knew what they wanted to say and did not need to take time for planning.
Contrastively, if the speakers did not know what to say next and needed
time to plan, the authors observed that they would use a filled pause rather
Goffman (1978) also addressed filled pauses as planning indicators.
He included them in his system of response cries,
where they functioned
mainly as place holders and signals of thought in progress.
As was mentioned above
, sometimes filled pauses for planning
have the added purpose of turn-taking management.
have presented descriptions of filled pauses for planning that could also be
interpreted as having the overlapping function as turn-keepers.
pauses in these cases are audible signals to the listeners that the speakers
wish to continue without being interrupted.
Schegloff (1987) explained that for many speakers,
complex ritual of coordination and anticipation.
conversation is a
He noted that as a person
a listener can make predictions about when the speaker will finish;
at that time the listener would be free to begin speaking.
There are times,
when a speaker might not want to let the listener speak,
the fact that she or he has more to say.
In this case, Schegloff noted that
a speaker could "modify the manner of his or her talk so as to circumvent,
ward off, or fight off such a start by another" (1987:208).
He explained that
one way he saw speakers doing this was by inserting a filled pause at a
sentence juncture or boundary where speaker turns often switch.
manner, the filled pause signals to the audience that the speaker wants to
Faerch and Kasper (1984) described similar language phenomena
They wrote that gambits
primary function [was] to maintain and regulate discourse" (69), such as
strategies used for turn-taking and turn-keeping.
of filled pauses as a subset of turn-taking gambits,
They classified the use
with a function similar
to that proposed by Schegloff.
hesitated and used filled pauses in order to catch the attention and gaze of
their interlocutors. In this case, rather than preventing the listener from
becoming the speaker, the filled pause signals to the listeners to redirect
their attention to the speaker.
Analysis of a Speech Sample
As we can see from
the scholars who
filled pauses have primarily looked at the planning function,
additional emphasis on those that could also be interpreted as turn-keepers.
The question that should be addressed is if this description matches the
functions of filled pauses as they appear in naturally occurring speech.
order to discover if the description is
samples to see how filled pauses are used by native and nonnative speakers
university student, and the other was a nonnative speaker of English who
came to the United States to study English.
In this transcript,
we see 17
9 used by the native speaker and 8 used by the nonnative
A look at the following exchange between a native and nonnative
speaker shows that the research so far has overlooked certain important
roles that filled pauses play in discourse.
By looking at naturally occurring
When researchers use a systematic analysis of discourse to
discover how language works, they are able to identify objectively patterns
Through this technique of analysis of natural speech data, we
can begin to see patterns in the use and distribution of filled pauses in the
speakers used a number of filled pauses.
(Refer to Appendix A for a full
key to transcription symbols used throughout this document.)
(1) A-1 M-Saudi Arabia/M-USA
1 > S. II like this uh people people who who married uh young.
2 I think that is the best way.
3 N. II think so too.
4 S. We say we say in our country, you safe to marriage very young.
-,,., S S,, ,,~,,,
S. I I didn't wanted to married from Germany because then you uh
uh. Ok I cannot stay in Germany.
N. Uh huh.
S. If you married in Germany or if you married in United States, you
have to stay in the United States.
S. Because if you go back to your country it could be difficult.
graduate my my undergraduate and then have to go to home and
S. It took time and time Germany was not too
year I have. Last year.
S. Since last year I am married.
I don't have any child. Do you have any child?
No, um we want to wait until I get into the Air Force.
S. Two years more, yeah?
> N. Urn it's
it's four years also.
Also. No I mean, when will you go to this military to?
when I graduate in two years basically.
In two years you will graduate.
You have no experience
You are now second year.
So you you need two years more.
Two more. Then I'll graduate and then I'll go into the Air Force.
But this time
I'll be an officer.
was in the Marine
Corps I was just an enlisted.
S. Ah yeah.
S. So then you'll have more money maybe and
> year left to graduate. And um, she'll come with me and she'll
try and get like a job on base.
S. What is her major?
> N. Um public affairs.
S. Public ((unclear))?
Oh public relations.
S. Public relations.
speaker were in clause-internal positions and functioned to provide
The next filled pause (in line 5) gave the speaker time to think
second filled pause in line 5,
in the clause-boundary position before
function as a planning hesitation.
in lines 8
Apparently, he was unable to say directly what he wanted,
so after the
for planning time,
he stopped and
approached the idea from a different direction.
mnsrried in n fnrpin rnmin.rv +th nnnnatf.ivP
Qnr lr~r lltaanA turn rnnra
or possibly he changed his sentence after the time taken to plan the
utterance (line 18).
These planning filled pauses that occurred in clause boundary
positions could also be considered to reflect turn-keeping gambits.
The nonnative speaker knew that he had something to say
pause sent that message
the native speaker to
appeared in a clause boundary pc
function of planning the sentence.
It appeared to have the
But also the topic of conversation
was potentially delicate: a personal question about children.
sudden change in topic into something delicate, we can imagine that
But looking at only one filled pause used in this manner gives
nonnrmative speaker asked the native speaker a series of questions
about his personal life.
In response to these questions,
speaker initiated many of his answers with filled pauses.
speaker volunteered to continue with an explanation of his plans, but
The filled pause in line 33 also appeared to be multifunctional.
First, it appeared to provide thinking or planning time and second,
it seemed also to soften a correction.
It appeared that the native
years to mean the full length of the ROTC program.
At this point,
he used the filled pause as a face-saving maneuver as he explained
that it was a four-year program.
examined using this multifunctional analysis.
Each one could be
identified as creating time for thinking or for planning. Additionally,
each could be
interpreted as having a softening function as
the nonnative speaker was asking very personal questions,
reluctance to discuss the money and responsibilities of a new job.
In this conversation, immediately after this question,
the nonnative speaker asked another question, this time about the
speaker used three filled pauses in explaining what she was doing
and the native speaker was apparently signaling some reluctance to
discuss his personal life.
The nonnative speaker
unaware of this reluctance, or this polite signal of reluctance,
continued with another question about the wife's
filled pause prefaced the answer (line 55).
The final filled pause (line
57) appeared to be a planning marker to give the native speaker time
to plan a correction to the title of his wife's major.
From a purely syntactic level,
we can identify placement of filled
pauses, but this does not give us much of a clue about their purpose or
function. A more sophisticated analysis reveals the use of filled pauses for
boundaries, filled pauses
provide time to
while clause-internal filled pauses provide time simply to plan a
word or other smaller unit.
Many filled pauses may be understood with
but there are others that require further insight.
analysis of a speech sample helps to demonstrate that filled pauses are
While they are being used to provide planning time,
pauses at claus
boundaries may have the added function of keeping a
Additionally, some filled pauses may function as softeners,
attempts to soften a potentially face-threatening utterance.
can be made by examining other components of the conversation, primarily
the subject or topic of
The Search for Other Functions of Filled Pauses
This examination of a conversation between a native and nonnative
speaker of English indicates that the literature on filled pauses gives an
incomplete view of the roles that filled pauses play in English.
analysis hints that filled pauses should be considered to be multifaceted,
with a more complex interpretation than had been previously suggested.
From what we have seen thus far
, we can identify functions of filled
on the context of use.
basic contextual interpretation
reveals the primary or most usual function of the filled pause, and that is
to buy planning time while speaking.
Filled pauses occur in both
Some of the filled pauses that
occur in clause boundary positions have the added role of managing turn-
In turn-taking, filled pauses provide time to plan, but they have
the added function of signalling to the listener that more speech is to come.
For the earlier researchers
However, as we could see from the example above,
there were some
positions that were signalling something else.
The excerpt above indicated
softening or signaling reluctance to speak. To understand this type of
semantic context; the subject or topic of conversation.
By doing this,
able to understand more about the less obvious functions of filled pauses.
we learn w]
id and use?
of filled pauses that native
way is to augment our traditional
analysis of conversation with the information that can be gathered from
This kind of interview allows us to tap into the
tacit knowledge about filled pauses that native speakers enjoy.
understand about their use of this particular aspect of speech.
ethnographic interviews (also known as
out loud protocols")
interviews in which native speakers are gently guided through a topic of
aspect of language.
These explorations reveal the implicit knowledge that
native speakers have and unconsciously use in their speech.
Spradley (1979) explained that an ethnographic interview could be
used to understand nonobvious elements of a person's life experiences.
, interviews were used for the purpose of discovering what one
group of native speakers knew about filled pauses.
Due to the fact that
filled pauses are generally not consciously used, ethnographic interviews
traditionally collected data of conversational dyads.
Outline of Subseauent Chapters
There are two principal aims of this research.
The first is to look at
speech samples of native and normative speakers to see how filled pauses
are used by each group.
This includes an analysis of the similarities and
differences in use between the two groups.
The second aim is to examine
data collection and analysis:
naturally occurring conversation and 2) the ethnographic interview.
combination to reveal an important speech phenomenon.
chapters of this report are arranged as follows:
of the data
ethnographic interviews of native speakers.
This information is compared
to that given by the earlier researchers presented in this first chapter.
the basis of the analyses of the functions of filled pauses that emerge from
the combination of earlier research and the native speaker insights revealed
the ethnographic interviews,
we are prepared for a more accurate
identification of filled pauses from naturally occurring speech data. The
analysis of this speech data is presented in Chapters Three and Four.
Chapter Three specifically explores filled pauses
in the speech
the subset of planning, managing turn-taking.
shows us that both native and nonnative speakers use filled pauses for
In the data that was collected,
we can also identify
some uses of filled pauses as planning that have the added function of turn-
the claims of
earlier researchers who had described these functions of filled pauses.
Chapter Four, however,
of filled pauses.
presents new information within the study
Based on the information provided by the informants from
presented above in Chapter One,
we can conclude that there is an added
In Chapter Four we see examples of filled pauses that
have been identified as functioning in this manner.
Also of interest is the
realization that primarily only native speakers used filled pauses with this
Filled pauses that could also be identified as softeners were much
less frequent in the speech of the nonnrmative speakers in this study.
In Chapter Five we will see a brief summary of the findings of the
ethnographic interviews and analyses of
This leads to a
numerical analysis for the quantification of filled pauses as they occurred
in the speech data.
From this analysis, we can see striking differences in
in the speech
of native and nonnative
implications of the study
as a Second
both in research and in classroom instruction.
Ethnographic interviews can provide a researcher with information
that cannot be tapped with more traditional experimental methods.
chapter presents the results of
ethnographic interviews with ten native
speakers, indicating that there are three overlapping functions of filled
pauses. In addition to corroborating the two functions described in the
planning and also in managing turn-taking (which overlaps with planning),
the informants were able to identify a third use for filled pauses, to soften
a delicate or sensitive subject.
This function was noted to be very broad,
information, we arrive at a new
deeper understanding of filled pauses than
had been discussed before.
For the interviews
, ten informants were selected from a larger group
of native English speakers employed as conversation leaders at the English
Language Institute (ELI) at the University of Florida.
The interviewees were all undergraduate or graduate students at the
time, from one month to more than two years.
The ages of the informants
were male. Nin
19 and 30.
Six of the informants were female, and four
of the ten informants were conversation leaders,
tenth was their supervisor.
For the analysis in this study, each informant
was assigned a number randomly from one to ten and is identified only by
The ethnographic interviews were conducted by the researcher in
classrooms or offices at the
ELI, and in
homes of the
At the beginning of each interview, several minutes were spent
putting the informants at ease with casual questions about their job,
studies, or unrelated matters.
When it appeared that the informants felt
comfortable with being tape-recorded, the researcher first read a paragraph
aloud explaining the nature of the information needed and then asked a
The informants were encouraged to explore their initial
responses as fully as possible,
after which more specific prompts were given
to help guide the discussion to find out what they knew about more precise
The interviews lasted an average of
The introductory paragraph and lead-in question were as follows:
I'm interested in the reasons why people hesitate while speaking.
You may have
that the speech
always smoothly fluent, and that at times people need to slow down
their speech in order to think.
There are different ways that people
can hesitate, for example, adding words such as "like" or "you know."
A YI n 4-l ^t nmn*nrn o + nrn l nn j +1c +n in rn nln a y nt +rii cina? nf nrnr
make noises, such as "umrn" or "uh."
In what situations would you use
an "um" or an "uh"?
From this general question, informants began to discuss what they
perceived as the role of filled pauses
in their speech.
provided by these informants that is relevant to understanding more about
(Transcripts of the interviews can be seen in full in Appendix B.)
This chapter is divided into two parts.
Part One presents what the
informants said in the first minutes of the interviews.
In each interview,
the researcher began by eliciting information about the uses of filled pauses
without giving any clues about the specific nature of the research.
the explicit knowledge
ten native speakers
about the use of filled pauses in American English.
The purpose of the first
general question was to get as much information about what the speaker
evidence that these native speakers could isolate and identify a variety of
functions of the filled pause.
The initial responses were comprehensive,
providing corroborating evidence for the uses of filled pauses identified by
additional function of filled pause use (softening), the second half of the
interview was aimed at collecting more specific information about subjects
that appeared to be delicate or sensitive.
In order to accomplish this, a
series of situations were brought up
in each interview.
were asked specific questions about situations that the researcher saw as
potentially delicate or sensitive, suggesting the possibility for the need of
softening or politeness.
The informants were asked to discuss how they
might react in each situation and if filled pauses would play a role in what
they had to say.
This information is presented in Part Two.
Part One: Defining a Three-Function Analysis of Filled Pauses
.-*-- --e 1s^t1 11 Tr e t io r *s 1 1 1 ** j1 11
in Part One
minutes of the interviews in response to the first general question:
as a basic
function of most filled pauses.
The second subsection defines filled pauses
that have the added task of manipulating turn-taking.
The final subsection
presents evidence about the hypothesis that the filled pause is used as a
softener for the sake of politeness.
Of the ten informants interviewed, six discussed the role of filled
pauses as planning hesitations.
These informants examined planning filled
pauses from many points of view, from how a speaker gains time to plan
the best words, sentences or topics, or in responding to questions.
following responses showed the speakers' understanding of the complexity
of this type of filled pauses.
Where and why do you use 'ums'?
When I have a break in my thoughts. When I'm trying to say
something and I forget my train of thought, I go 'um' to give me a
second to register in my mind what I'm trying to say. And if I'm
talking to someone else, it gives them a second to figure out what I
have just said. So there are different uses for it. I think it serves
two purposes. It gives the speaker time to re-think what they're
saying, and it gives the listener time to acknowledge what has been
What do you use 'ums' for?
When you are just stopping and thinking.
a question and you haven't said anything.
Like if someone asks you
I guess just stopping and
Why do people use 'um' when they talk?
To think of a better way
mind. You have all these
them all, but you have to
you use a hesitation to get
of saying the next thought that's on your
things in your head, and you want to say
slow down and think of an order, and so
everything situated and then have it come
Where do you use 'um'?
When you don't know exactly what you're going to say.
to collect your thoughts.
What uses can you think of for 'ums'?
You're stalling for time; you need to think, to come
something better to say, or you can't find the word that yo
use. You're searching for something, and it's going to take
time. But I think generally it's just stalling. That's what I
u want to
Stalling as in planning?
Slowing things down.
In what situations in conversation do you use 'urns'?
I've noticed a lot when you're speaking in public. You take that
moment to get your thoughts collected. Maybe when you're at a loss
for words. When you find yourself not being able to say that word
that you know, the one word that fits there. So you go 'uh' and kind
of think about it. Also when you have a lack of words. And you
don't know what to say but you don't want to come out looking
stupid, so a big moment of silence ends your conversation, and if you
to maybe guide
I your point, you look kind of stupid, so you use 'uh'
As we can see from these responses, these six native speakers tell us
that filled pauses are used for what they colloquially call "gathering their
The literature cited in the first chapter revealed that planning
filled pauses could occur in sentence or clause boundary positions or in
of planning filled pauses for what they called planning for words (clause
internal) and planning for sentences (clause boundary).
second overlapping function of filled pauses was in turn manipulation. Five
of the informants explained how filled pauses used as planning also had
this added role of letting the listeners know that more speech was to come.
The following excerpts illustrate their understanding of this special role of
the planning filled pause.
What uses of um'
do you know of?
It's a pause.
It's definitely a pause.
To maintain the right to speak.
To keep the next person from interpreting the silence as a point
where a decision is made to take a turn.
I'm not done
It's a signal.
going to continue speaking.
It means that
[ just haven't
haven't relinquished my turn,
while I'm trying to formulate what I'm going to
that something is
. It's a signal to
When do you use
When you're speaking and you break, that gives someone a pause to
I guess the 'um' is to continue,
I'm not finished.
It's just a filler for you to say
What's the difference between using an
'um' and a silent pause?
getting your thoughts.
You're not just stopping and you cut them off.
It lets them know that you're still going on with what you're saying.
What uses can you think of for
Maybe it's used just to keep attention. As long as there
are sounds coming out of your mouth, they'll be paying attention.
you become silent too long, you might lose their attention.
And then what might happen?
Why do you 'unm'
rather that pause silently?
Maybe to keep people's attention.
So they know you're about to talk.
Kind of like holding your thought in the conversation.
say something, so don't interrupt me."
"I'm about to
interpretations of this type of filled pause.
The first three informants noted
the two-sided nature of filled pauses used in this way: 1) the filled pauses
are used to
to plan the next utterance; and 2) they are also
signals to the listener to not begin speaking.
This is the same insight as
that given by the researchers presented in the first chapter.
insight or interpretation of filled pauses used in turn-taking management.
role of the
in keeping the
revealed until now.
Softening for Politeness
The ethnographic interviews uncovered a third major function, that
researchers have overlooked.
In addition to planning and managing turn-
taking, the interviewees noted that filled pauses often had the added role
of softening certain difficult or sensitive situations in native speech.
responding to the
initial prompt, seven of the ten informants discussed
filled pauses as signals of politeness when talking about sensitive subjects.
These examples show us the control that native speakers have over an
element of language that has varied uses.
Where else might you use an 'um'?
I guess also when you try to interrupt somebody.
something to me and I'd say "Uh::
You can be saying
You know I don't agree with you."
In other words, I
don't just say "I
don't agree with
So maybe I would slowly interject into what you're
What role does the 'um'
play in this situation?
I think maybe to be polite.
I think you come off a lot stronger and
saying "NO!" which is kind of rude.
When do you use
In that sense it plays a social
is usually used to fill a spot when you are thinking about what
you want to say.
You don't want to put your foot in your mouth,
especially if you're talking about a sensitive subject.
, as you're
talking, instead of giving somebody a pregnant pause, you fill it with
something because you're speaking.
Any other situations?
When you're not sure if you want to say something,
going to upset the other
person, you kind of "Well,
This last informant (#7) brought up the word hedging to identify how
she saw filled pauses functioning in this manner.
Hedges may be defined
as verbal fences that help separate a speaker from a difficult subject,
other words softening devices.
In many cases where filled pauses are used
we can also consider them to be hedges.
This is examined in
more detail in Part Two of this chapter.
Do you ever use 'um'
when you're thinking about how to say something?
If I am in a situation where I have to put some
kind of information a certain way, so I won't hurt someone's
then I'll say 'um.
Because I don't know how to say it.
I don't know
r .1 1 I .r
If you 'um' before giving an answer, is that better in some way?
, I don't think it has anything to do with being offensive or not.
I think it's
just a stall that I'll use.
I think it shows people that I'm
about something too.
But normally, I think if I say "Um,
well," it cues people
that maybe the answer is not what they want to hear.
It builds it up
and gives me time and at the same time it gives them a little clue
that maybe the answer isn't going to be what they want to hear.
exactly what it is.
One of these speakers also noted that filled pauses used in this type
of situation could be interpreted as revealing dishonesty.
Any other thoughts on 'urns'?
sometimes if they're being dishonest.
I think if you can't say it right
away, you must be searching for something to say.
Finally, there were two speakers who addressed how the listener
might respond to a filled
pause used to reflect politeness,
unconscious awareness of a simple filled pause.
Their answers indicated
that their responses to a speaker might be influenced greatly simply by the
use of an 'um.
Do you think the
'um' makes it easier for the other person to hear that
Possibly, because they know you're struggling to say something. And
if you come out and just say something,
direct or you don't care.
So does the
they may think you're too
'um' soften it?
Yeah, it could soften the blow.
I definitely think that.
As a listener
if someone answered
" how do you respond.
I would take that as a "NO."
Or as something that's
not definite at
I'd perceive it as something that's questionable and probably will
be not what I want to hear.
If they're hemming and hawing right
If I hear it
, I think something not good is going to come out
Do you prefer a filler or no filler?
With a native speaker, I prefer to be cushioned.
So what do you think about just saying "No."
I'd like to have that cushion, that 'um,
' or at least to
make it seem like they thought about it a little bit; prepared me for
the let-down, then let me down.
Three informants who mentioned the need to use filled pauses with
sensitive subjects also noted a differentiation between close friends and
Based on what these informants said, the need to be overly
polite would not exist among close friends.
They also noted, however, that
i1 .I Ia -i 1 1 i 1 1 1 A
Can you say more about sensitive subjects?
It depends on how well I know them.
If it is a close friend, I'd just
come out and say what there is to say, but if it's someone
know that well,
I have to be careful.
You don't want to upset or
For you, using an 'um'
is part of being careful?
, a part of being careful.
What is different about using a silent pause instead of filled pause?
I think it would be easier for people to lose interest or get distracted.
think it has
something to do
are in your
(indicating that in close relationships,
fewer filled pauses are used to fill in awkward moments),
What is it about an uncomfortable situation that would make you use an
Just the pauses maybe.
you're going to say.
You have to think more before you say what
So while you're thinking, you're saying 'um.
would depend on how well I knew the person.
It really depends on
the relationship with the person.
That the use of filled pauses also reflects the closeness of a social
relationship makes sense.
If social distance plays a role in pronoun use,
1978; and West and Zimmerman 1985)
use of filled pauses in politeness. To (
it should also influence a speaker's
confirm this initial analysis of filled
pauses in softening for politeness, more
information is needed.
presented in the following section.
Part Two: Exploring Three Samples
In order to elicit specific responses about filled pauses as used to
soften for politeness, three scenarios were introduced, and the informants
obvious functional use of filled pauses.
By focusing their attention to filled
pause use in these highly specific scenarios, the informants were able to
produce valuable information about their internal, tacit knowledge of this
specialized language function.
The scenarios presented in the following three subsections are
speaking about body odor to nonnative students and to American friends;
2.) asking for a favor; and 3.) asking personal questions, such as age.
were based on actual situations that the subjects may have had to face.
Situation One: Discussing Body Odor
The first of the scenarios involved a problem that had arisen at the
English Language Institute about the body odor of several of the students.
This subject came up because it was the responsibility of the conversation
leaders to discuss the topics of health and personal hygiene and body odor
in their meetings with students.
Clearly, this topic clearly represented a
researcher simply asked each informant how she or he would use filled
pauses when addressing this topic in classes with nonnative speakers.
Eight of the ten informants addressed this subject and indicated that
One informant explained what he believed to be the role of filled
pauses in sensitive subjects in general.
When someone gets flustered.
When the subject of the conversation
falls onto a sensitive subject, lots of
start coming out.
they're reluctant to talk about something.
The following responses were given to the question,
"If you have to
speak to a student about body odor
would you use any
Would you use any 'ums'?
, I'd be 'umming' all over the place.
It's just going to be a big 'um'
situation because you don't want to insult them.
have to "Uh,
to see that we are struggling
So we're going to
So hopefully it will give them a chance
. So hopefully if I'm stuttering, maybe
they'll see that I'm uncomfortable and they'll explain it to me.
From a slightly different perspective came the next answer.
informant's primary concern was for the reaction of the listeners. I
the role of the filled pauses as giving him time to examine and respond to
Would you use
Oh yeah. Quite a few.
Anytime that I've broached a sensitive subject like this,
I always use
the pause to try and judge how the person is taking what I'm saying
and get a feel for how offended they are, so that I can direct what
I'm saying in a way that they'll understand without being blocked by
anger or resentment.
So I'll use that pause to let them know that
they can speak if they want to.
The next answer reflected the speaker's concern for being able to
have the best way to word what he had to say in this kind of uncomfortable
He indicated that he would use a filled pause to create time to
think about what needed to be said.
This response blurred the line between
planning and politeness, indicating again the overlapping functions.
speaker indicated, though,
that the planning time would be used to think
This informant also noted that there would be a difference in
how he would act with different people with whom he had varying degrees
Would you use an 'urn'?
That would be an uncomfortable situation.
What is it about an uncomfortable situation that would make you use an
Just the pauses maybe. You have to think more before you say what
you're going to say. So while you're thinking, you're saying 'urn.' It
would depend on how well I knew the person. It really depends on
the relationship with the person.
Similarly, the next two informants discussed the need to plan their
statements carefully in any kind of awkward situation.
The first informant
about the reaction
and how that would
influence how she would say what was necessary.
Would you use any 'urns'?
Awkward. Because I know I'm trying to get a message across to a
specific person. And I want to figure out if they're getting the
message or not. And it's an embarrassing topic for an American to
talk about because we don't talk about body care. So I think I'd be
a little embarrassed to talk about it to a bunch of strangers.
So when you're embarrassed, you use 'urns.'?
I probably do, now that you make me think about it.
Well, because if you're shy, like I tend to be, I don't like to be direct.
If I feel any discomfort, then I would probably pause in my speech,
or if someone asked me something that I was uncomfortable with, I
'um' make a difficult situation easier?
It gives me time to register everything.
Would you use an 'um'?
I think I
And definitely when I start to present the
it's something that we don't feel socially comfortable talking
Not only body
but in Level
You're kind of
So you go
' you know, hesitant.
, maybe when you feel uncomfortable talking about something.
Just like body odor.
You don't want to be offensive, so you're maybe
using it to hold on a minute and run it through your head before you
say it so it doesn't come out sounding wrong.
I guess in that sense
you'd use it in situations where you want to make sure you don't
come off sounding the wrong way.
a lot this week.
I've been using 'um'
Does it make it easier using an
It makes it easier for me.
delicateness of the situation, combined with the discomfort for her and the
potential embarrassment of everyone involved.
When asked if she had any
provided an explanation of how native speaking listeners would interpret
the use of filled pauses.
This indicated that she was aware that native
speakers understood that filled pauses provided signals to the audience that
something difficult was happening.
How would you feel in this situation?
That would be really delicate for me.
I wouldn't feel comfortable.
Would you use 'um' at all?
Especially if I knew that student was sitting next to me.
Why would you 'um'?
I don't know
they don't kn
if they would
if you know you're embarrassing someone, because if
ow that they have an odor problem. I don't even know
think that was directed at them. Or maybe do it on
glance at them, and maybe they would know.
How might you use an 'um' in this situation?
Because I'm trying to be
You're trying to phrase it
hard to get the exact style
about how best to say it.
delicate and not embarrass somebody.
delicately to get the message across. It's
you want. You're just pausing. Thinking
It's a thinking place.
I don't know
or if you're
"Oh, you are
if you're using the 'um' as a filler while you're thinking,
using it like an indicator to the person. Using it
or undeliberately because you've just learned and you
that you've learned that that is a way that people say,
obviously hesitating, what's wrong, what are you trying
to say." Because if people say 'um' too much, and people say,
ahead, spit it out."
the next passage,
the speaker stressed the need for the filled
to provide an element of politeness in a situation
that could be
interpreted as offensive.
How would you react in this situation?
You have to be very polite about it.
Do you think urnsn'
play a role in politeness?
Because you're trying not to offend anybody.
I would think
of it as a form of hedging, because you're saying what you want to
say, but you're hedging from saying it directly.
Putting it in a more
She indicated that with the nonnative speakers she
would not hesitate to discuss the problem.
But, if the situation involved
she would behave differently.
This indicated her belief
that while native speakers were aware of softening strategies in politeness,
nonnative speakers did not necessarily have that same awareness.
other hand, seven of the informants indicated that they would use filled
Would you use
With nonnative speakers?
right out and say it.
But if I
approached an American
student about it, I would definitely 'um'
would do that.
wouldn't just say,
This collection of examples illustrates that the native speakers feel
that filled pauses played an important role for them in softening to achieve
Situation Two: Asking for a Favor
The second situation presented to the informants was the problem
of asking for a favor.
Due to the fact that several of the informants had
previously noted a difference in politeness strategies between close friends
and more distant acquaintances,
the situation for discussion was described
asked how they would ask a favor of
someone they were acquainted to,
but not very close to.
For many, asking
others for favors was a very difficult situation to be in, and for this reason,
several informants simply stated that they would not ask an acquaintance.
One informant preferred to walk home five miles late at night in the rain
than to call an acquaintance and ask for a ride home.
Those who were willing to consider the possibility of asking a favor
came to the realization that 'ums'
could be used as signals to the audience
that something besides a favor was being asked.
This type of politeness
"saving face" by
Brown and Levinson (1978).
authors write that in a situation that the speaker is making himself or
speaker may do something to "save face," to protect himself or herself from
asking favors may be a form of
a face saving strategy.
explained it like this:
What would you do in this situation?
I would probably begin with an 'um.
' Or some other signal like that.
What would it be a signal of?
"I'm about to do something I
It's going to involve you!"
"I'm about to expose myself,
so please be careful."
Does the 'ur' make it easier?
I don't know.
It's like a signal.
If you say 'um,
' I know you're about
to weasel into something or try to weasel me into something.
Any other uses of 'um'?
It can be used as a softener.
When you're asking a favor, it can be
used to soften the fact that you're making an imposition on the other
Or to let the other person know that you recognize that this
could be perceived as an imposition.
Or to deliver unpleasantness.
Another informant also perceived the role of the filled pause as a
signal to the listener.
In the following passage, he addressed how the use
um' would affect others in general,
and himself specifically.
In asking a favor
would you use an 'um'?
came up to him and realized at the last second that
maybe I shouldn't be asking this person, the 'um' would be there.
think that kind of
'um' conveys a sense that they're hesitant to ask.
.wn I tiyr
going to take up some of my time or that I'm going to have to do
The following segment addressed again the issue of how the use of
filled pauses for softening depended on the closeness of the relationship.
This informant explained that when asking a favor of a close friend, there
in this manner
Although she suggested that she did not use filled pauses
, she indicated that she had been aware of others using this
When you ask a favor,
would you use an 'um'
with a close friend?
No, not unless it would put them out. And then I would probably
wait for a long time until I could just say it. But my roommates do
I've heard them put them in there.
Asking a favor could potentially be face-threatening, and several of
the informants recognized the added softening characteristics of the filled
pause in these cases.
One informant explained how this worked for her.
If you have to ask someone to do a favor that they may not want to do,
would you use an 'um'?
It just softens the blow.
It's just like in the other case, like when
you're telling them something that they don't want to hear. It's
almost the same as saying that you know that they possibly won't do
it or they might not want to do it. But they might. You never know.
It depends on the way you say it, maybe. It could soften the blow,
and you might get a different reaction. Maybe they'll pick up on you
wanting to ask this favor, knowing that it's difficult for you to ask it.
So you're 'umming.'
Do you think it's more polite?
In a way it is more polite.
ask a favor, I don't come o
this for me." I don't talk
y'know, y'know, y'know."
Because I find myself whenever I have to
ut and say, "Oh, do you think you can do
like that. I say, "Y'know, do you think,
As a recipient of a request for a favor, are you more accepting if a person
Yeah. Yeah, I think so. It's almost like you're saying to them
without saying anything that it's really hard to ask. And I think
they're going to be more receptive. I don't think it's conscious.
People don't say 'ums' because they know they're going to get a
favorable response. I think they just do it naturally.
One other informant tapped into his native speaker knowledge of the
emotional impact of using a filled pause with a request for a favor.
recognized his own use of this politeness strategy to convey that the request
might not be well received by the listener.
Would you use an 'um' in asking a favor?
know you feel uncomfortable about having to ask them.
So I think
To show the person that I know I'm putting them out.
As a recipient of a request, which is more polite, with an 'urn'
I'd rather receive it with that emotion, that filler. It le
ts you know
that the other person
You feel like
it down a
, "Yeah, I'm going to help this person."
they're showing you that they are not really comfortable with what
So yeah, I think if somebody asked me without it and
somebody asked me with it, it would make a difference in the way I
In the following two segments,
the insight was put forth that the
filled pause in asking personal questions was used as a planning strategy.
For these two speakers,
the purpose of the filled pause was to
allow extra time to plan in order to make the next utterance, the asking of
, more acceptable.
This first informant indicated that she did not
like to manipulate people,
and she needed the extra time provided by the
filled pause to plan for the most strategic and least offensive way to get the
Would you use an 'um'?
Yeah of course, because I don't know them so well, and I go to ask
them the question, I'm way out of line for asking the question in the
They have no obligation to me.
And I'm desperate.
Does the 'um'
make it easier for you?
manipulate this person.
'urn' sounds nicer.
It doesn't sound as
When you use an 'urn' before it, then you're going to soften
and soften it, it's not as forward and
direct, and it gives you more of a chance to say no.
so direct, it puts the person I'm asking in shock. I
If I just ask, it's
[t doesn't prepare
them for what I'm going to ask.
So I thing the
'um' kind of helps.
The following speaker expressed a similar desire to plan ahead to
give herself time to make the request look appealing.
How would you ask a favor?
Beat around the bush.
Do you think you would use any 'ums'?
probably would because
I'm going to have to
constantly be thinking of things,
to make it look enticing.
The responses by these last two informants were different from
They provided yet another insight into how these native
speakers perceived their language.
While some indicated that a filled pause
used in this manner was a signal to the listener or a marker of politeness,
other native speakers interpreted this behavior as planning time for the
speaker to find the least offensive way to ask the question or favor.
nI ~ I*~n n~rr Clr n C n m I I i I; C1 IH nlj nill n i nH nln:n Wl~nr: r] nn 1:C1\ nW :nCn~nYlnCn~:nH
Situation Three: Responding to Personal Questions
The final situation presented to the informants was the problem of
such as age,
income, or weight, and were asked about how
they might answer such a question.
The situation that provoked the most
illustrative examples was
a person asked a
woman her age,
simply answered with a filled pause.
The informants were asked how they
would interpret the response.
communicative competence in the use of filled pauses.
She described the
type of person she would expect to violate the unspoken rules of using a
filled pause in response to a personal question.
If you receive a response of only 'um'
too a question,
how do you react?
Even here if you're a native speaker, but just a very self-centered,
unobservant native speaker,
you still might say,
Most people would realize,
The following responses came
three of the
They appeared to have a more defined idea of the delicateness
of the situation than the female informants.
These males indicated that
they would interpret a filled pause as a definite sign of reluctance from a
When you ask a woman,
would you do?
"How old are you?" and she answers "Um::" What
would feel like I asked something that was,
that I was too early.
speaking about certain things.
If I had asked somebody "How old
and instead of saying,
, I'm 29," they
would feel uncomfortable with my question.
What would you do then?
I wouldn't push it.
I wouldn't ask again, no, or not necessarily the
age thing, but if you were asking someone out on a date,
"Mm::," you know immediately that's
If you receive an
um' as a response to a question,
and they go
how do you react?
I would probably give it a second, and if they didn't come right back,
Respect their right to privacy.
They're giving me that gap,
that lapse of language so I can insert my apology.
the 'um' being used as reluctance
Yeah. in that context.
How would you interpret an 'um'
as a response to a personal question?
If I ask a woman her age,
and she tells me right away, it's
not a big
issue for her.
But if an
obviously an issue for
on the other
h\^"il O1 O +t1o T oT n rn on onnn,'nl'nrJ n nf^t L* nsa n nr4 n--^4-l t11 n.Jl ^
see that for this
is a difference
between males and females with regard to the use of filled pauses and their
Summary of Ethnographic Responses
The passages presented here showed that the native speakers who
The ten informants discussed aspects of filled pauses in planning,
managing turn-taking, and even in softening difficult situations.
responses corroborate what other researchers have said about planning and
turn-taking, and they also provided insight that filled pauses had a third
use in softening
In order to
how these insights match
filled pause use in naturally occurring speech, Chapters Three and Four
present an analysis of spontaneous discourse between native and nonnative
FILLED PAUSES AS PLANNING IN
NATURALLY OCCURRING CONVERSATIONS
As we have seen, researchers have identified planning as the primary
function of filled pauses.
We have seen corroboration from native speakers
particularly in protecting the speaker's
right to a turn.
examples of naturally occurring filled pauses from the speech of both native
and nonnative speakers.
Speech samples from both native and normnnative
speakers were analyzed because it is important to understand the functions
of filled pauses for both groups of speakers.
An analysis of native speech
alone is informative, but it cannot tell us what nonnative speakers do with
may provide us with
miscommunication in conversations between speakers across cultures.
The chapter is presented in the following format:
of the procedures for the collection of
presented in two main sections. The fir
that occur in clause-boundary positions.
After a description
data, examples from the data are
'st examines planning filled pauses
After a general discussion of filled
subdivisions of clause boundary filled pauses used for planning.
functions that emerged from this analysis of speech.
The first is in topic
switching and the second is in side sequences and interruptions.
presentation of examples explaining each of these, there will be a discussion
In order to examine the use of filled pauses in native and normnnative
eighteen pairs of subjects were tape recorded.
consisted of one nonnative speaker and one native speaker
for a total of
The native and nonnative speakers were selected based on
The native speaker subjects were all students at the University of
Florida and were enrolled in an introductory linguistics class at the time
English speaking citizens of the
They were selected as
partner program at the English Language Institute (ELI) at the University
of Florida, from which the nonnative speakers were selected.
The ELI is an intensive English language program for international
, affiliated with the University of Florida.
Many of the students
beginning to high intermediate/advanced levels.
placed into one of six levels,
Students at the ELI are
and are initially assigned to these levels based
a writing test, and the Michigan test of language
The students take classes in reading,
skills, and they also have a small selection of optional classes available to
them, including a TOEFL preparation course and different conversation
including the conversation partner program.
This program pairs
ELI students who desire more conversational contact with native-English-
instruction in the students'
The nonnative speaking subjects were students enrolled in the ELI.
willingness to participate in the conversation partner program.
in order to test a group of nonnative speakers at similar levels of language
development, subjects were selected from available students in Level 4A
and Level 4B
, levels of high-intermediate English proficiency.
their placement in the levels by the three evaluation tests, an assumption
was made that the subjects represented a reasonably homogeneous group
in spoken English ability.
All of the subjects had taken some classroom
instruction in English in their home countries prior to coming to the United
As can be seen in Table 1
, the students represented a wide variety
of language backgrounds.
approximately 30 minutes of one of their conversations.
took place in many locations, ranging from the lounge at the ELI to local
speakers were given tapes by the researcher and were instructed to record
arrangements to record the conversations using a variety of tape recorders
Nonnative Speaker Subject Information
A Saudi Arabia 4B M M
B Indonesia 4B F F
C Taiwan 4B F F
D Panama 4B M M
E Brazil 4B F F
F Turkey 4B M M
G Venezuela 4B F F
H Taiwan 4B F F
I Taiwan 4A M M
J Japan 4A F F
K Venezuela 4B F F
L Spain 4A F F
M Taiwan 4B F F
N Taiwan 4A F F
0 Taiwan 4A M M
P Korea 4A M F
Q Spain 4A M F
R Korea 4A F F
that were available to them, from portable microcassette recorders to larger
stereo systems to telephone answering machines.
The quality of the tapes
varied, but each was clear enough to allow an accurate transcript to be
coached in any way by the researcher.
The audiotapes were returned to the
10 minutes were transcribed and checked for accuracy. In
transcribing the tapes, an attempt was made
to record every word and
filled pause that was spoken.
Methods of Analysis
Researchers have successfully used analysis of conversation to reveal
patterns in language that are not readily noticeable to casual observers.
researcher can make conclusions about subtle uses that occur somewhat
beyond the consciousness of the average language user.
Although they use
these language patterns every day, they are unable to explicitly describe
what they do.
By analyzing conversational discourse,
we can learn more
about the language in use around us.
For the current study, an analysis of filled pauses was conducted.
Filled pauses were identified as nonlexical sounds produced by speakers.
In order to narrow the search for those specific nonlexical noises that could
and transcripts of conversations among different groups of native English
There was a total of 85 filled pauses that were identified in these
66 transcribed segments.
There were two forms of filled pauses that were
used: the most common was "um" (with 50 out of 85 filled pauses),
second was "uh" (with 35 out of 85 filled pauses).
With this knowledge of the physical characteristics
the search for filled pauses in the collected data began.
transcribing the tape segments of each of the 18 pairs of speakers described
in the previous section, the analysis of discourse was accomplished by first
identifying each occurrence of the filled pause.
It quickly became apparent that although "um" and "uh" were very
there were more than these two pronunciations of filled pauses
used by native and nonnative speakers.
The speakers used the following
nonlexical sounds as filled pauses for hesitations: "um," "uh," "m," "em," and
Other nonlexical sounds that were used
, but not as filled pauses were
-agreement, affirmative response
-understanding or realization
-sound of distaste, dislike
-(rising intonation) questioning,
asking for repetition
-(falling intonation) noncommittal response
After all of the filled pauses were located,
they were identified as
being either in clause-boundary or clause-internal position.
context surrounding each filled pause was examined to discover if the filled
pause occurred in association with a delicate or sensitive topic.
this information, a decision was made regarding the use of the filled pause
as a hesitation,
or as a hesitation
identification of filled pauses came the following analysis.
As a reminder of where clause-boundary pauses could appear, in
Boomer (1965) reported that a position after the first word of
a clause was common as part of a false start to a sentence.
The first passage showed a nonnative speaker using a filled pause to
make time to plan his next clause.
The filled pause here gave the speaker
time to think about what he wanted to say as well as how to say it.
served the purpose of letting the native speaker know that the normative
containing the filled pause(s) being discussed.)
When you go to Taiwan.
Go Taiwan, and uh you face many many problems.
And you have to write down some sentence.
Oh. Ok. Like in this book here?
remember the details of an evening with her friends when they went out
to dinner and to a dance club (Kaos),
and the native speaker was trying to
The first filled pause in line 3 provided time for the speaker to
gather her thoughts and her grammar and to say the sentence.
Ok. So we ok so we so you can say um we were there yeah
you were right.
We were there for like two hours.
After that uh we were we went to the eh Kaos.
Another common place a clause-boundary planning pause appears is
when a speaker is having difficulties making his or her point clear.
, many native and nonnative speakers used a filled pause to plan their
idea or sentence.
The next few examples are presented to illustrate this
particular type of planning filled pause.
problematic sentence to convey the information more clearly.
speaker in the next excerpt was asking the nonnative speaker if he had
nonnative speaker did not make his idea clear to his partner and needed
to repeat the idea.
(3) D-1 M-Panama/M-USA
1 NS. ((laughter)) Haven't you uh I mean did you know a
2 little English before you got here or
That's all I know already.
NS. You didn't know it. Oh this you already
learned this already.
When I get here I was y'know
Empty. No English.
The native speaker misinterpreted the nonnative speaker's answer to the
This resulted in the nonnative speaker making an attempt
to revise his explanation to be more clear, with an "um" starting off his new
But even with this planning, it still took him two tries (lines 6
and 8) to get the information across.
In the following example, the nonnative speaker used filled pauses
while planning the second clause of the sentence.
to successfully create
However, he was unable
what he wanted, and he abandoned it in favor of
another, prefaced by "ok."
(4) A-i M-Saudi Arabia/M-USA
NN. I I didn't wanted to married from Germany because then you
uh uh. Ok. I cannot stay in Germany.
NS. Uh huh.
NN. If you married in Germany or if you married in the United
States, you have to stay in the United States
NN. because if you go back to your country it could be difficult.
NS. Yeah. Awkward.
the nonnative speaker appeared to think about,
then give up on,
After he got a running start on the idea by approaching it
from a different angle,
he was able to complete his thought.
In the following example,
the speaker did not know how to say what
After asking aloud and continuing the planning hesitation, he
abandoned the idea completely and tried it in a different way (line 6-7).
(5) Q-1 M-Spain/F-USA
NN. So so different.
Uh first in Europe the life is more expensive
Much more expensive than here.
So. And, because um it's
I like USA, but I like Spain too so.
NS. Uh huh.
um how to say some.
America I like
The filled pause in line 1 gave him time to plan his sentence,
and again in
, he needed two filled pauses for planning time.
Even though he took
In other c
the data, a clause-boundary filled pause was
combined with the phrase,
"I think," indicating that the pause was needed
for the planning of information that was not known for sure.
This was seen
interjections such as "uh" to signal to the listener that they were unsure of
the information they wanted to say.
In these next examples of nonnative speaker clause-boundary pauses,
speaker was thinking of the information needed to answer a question.
(6) H-l F-Taiwan/F-USA
Dh. How many people in each group?
Um I think four
, four, year
four only four
two three four. Becau- because we we have uh 17 17
students in one class.
In the cl-. Oh
So so so four in one group.
The nonnative speaker needed time to think about the correct information
to answer the native speaker'
The speaker possibly used the
phrase "I think" to indicate that she was doing more than simply looking
for a word.
In example (7) the nonnative speaker also used "I think" in addition
What was uh Jeff's
NS. Is he how long has he been here
I think three months I think.
Oh, that's all.
In this example,
the nonnative speaker said "I think" two times, indicating
that he may not have been completely confident with his answer.
language fluently, there is a chance that the speakers will not understand
presented here that many listeners use a filled pause at the onset of her or
his turn, providing time to understand the previous utterance. Rather than
directly indicating that they did not understand,
several speakers used this
option of using a filled pause followed by some kind of clarification question
next two examples show native speakers
misunderstanding the nonnative speakers.
Example (8) was how one native speaker used this strategy.
nonnative speaker was trying to explain why she had trouble with speaking
English fluently, and the native speaker was not able to understand what
she was saying.
The listener needed time to think about the information
she received and to plan an appropriate utterance in return.
(8) C-1 F-TaiwarF-USA
But II should learn how to say in Eng- English
And some um like
um uh a lot of sentence, I, uh I use the Chinese way.
NS. The order?
Ya, order, ya.
Uh ((unclear)) The how ((unclear))
It's different from Chinese sort of.
NS. Like the placement of words, and sentence or how you.
Is uh expression.
Ya, expression. Different,
very different from Chinese,
should learn how to express.
Yeah, expression self.
not appear to
speaker was trying to say here.
In addition to the filled pause in line 6, the
native speaker mumbled something that might have been her interpretation
of what her partner was saying, but this did not help her to understand, as
the suggestions in lines
The discussion continued
while they tried to negotiate the meaning,
with limited success.
result of what the
nonnative speaker feared:
that she was
using Chinese "expressions" instead of English expressions.
this next example,
the nonnative speaker was
speaker said specifically that she did not understand what the nonnative
The multiple utterances of wait plus the filled pause in line
4 gave this native speaker a great length of time to process the information,
apparently figured it out and cut off the nonnative speaker before she could
rephrase the troublesome utterance.
(9) G-1 F-Venezuela/F-USA
1 NN. and ah one of one of those? Ah one of them.
2 NS. One of them?
3 NN. Came here uh and stay with me like uh two weeks?
4 > NS. Wait wait wait wait wait uh I don't understand what you're
NS. One of your aunts stayed with you?
NN. Yes here.
Before the nonnative speaker could respond to the complaint in line 4, the
native speaker had enough time to think, and she interrupted in line 7 with
a rephrased version of what she thought she understood, which turned out
to be the correct idea.
Even though she said that she did not understand,
given enough time, she was able to process the information.
In this case
, the filled pause also serves as an indication to
speaker that there
comprehension of the
question often provide enough planning time for the nonnative speakers to
construct responses without waiting for the native speakers to restate or
explain. The nonnative speaker in example (10) did just that.
(10) R-1 F-KoreaF-USA
1 NS. About having a major? What do you wanna do?
2 > NN. Uh when I was in Korea? When I was in Korea I
studied art. For
The nonnative speaker in line 2 initiated her utterance with a filled pause,
followed by a clarification question,
with the answer to the native
listeners interpreted some filled pauses correctly as indications that their
nonnative speaking interlocutors were facing a troublesome utterance.
the following two examples, the nonnative speakers produced filled pauses
in ways that were interpreted by the listeners as indicators of problems
partners then provided the nonnative
speakers with explanations.
(11) N-l F-Taiwan/F-USA
You know what work out.
(12) R-1 F-KoreaF-USA
Like right. And over here is Criser Hall.
I'll I'll draw you I'll show you sometime.
In these two cases, the filled pauses apparently elicited further explanation
by the native speakers. It was possible that these filled pauses were simply
clause-boundary pauses to give time to think about what to say next, but
the native speakers provided more information to their nonnative speaking
category of clause-boundary filled pauses used as planning.
somewhat different functions of the
pause is to provide an overall, comprehensive picture of some of the related
uses of this kind of filled pause in a clause-boundary position.
there are slight differences in the descriptions and grouningss of these first
examples, the uses are similar enough to suggest a single category with
subtle variations in use, that both native and nonnative speakers use.
The following three subsections describe functions of filled pauses
sufficiently different to be presented individually as subdivisions of filled
pauses used as planning.
The first group to be examined is planning filled
pauses with the added function of turn management,
which we have seen
The second subtype to be examined is in topic switching.
data examined, filled pauses frequently occurred at the initiation of a topic
division of clause-boundary filled pauses that emerged from the analysis
or interruptions of
speakers would use a filled pause upon returning to a topic of discussion
or a secondary
conversation within a conversation.
As we have seen, clause-boundary filled pauses have been noted as
helping with turn-taking management.
a sentence or clause boundary, the lis
If a speaker uses a filled pause at
itener frequently knows that more
boundary to begin speaking, the listener might recognize the filled pause
conversation were identified to illustrate control and lack of control of this
aspect of planning filled pauses.
(13) D-1 M-Panama/M-USA
You live with by yourself there?
No! I have three roommates.
Two American and one
Yeah, one of. He also is learning Spanish ey
NS. English, yeah?
NS. And Spanish probably.
Uh huh, yeah,
well just but words,
you know what I
((laughter)) I know, cause,
, go ahead.
like David David, teach me teach me some
What is called y'know.
indication of planning,
and he allowed the nonnative speaker to continue
with his story.
Although it was expected that native speakers would be aware of
these sentence planning hesitation markers when they were used by other
this was not always the case.
Example (14) showed a native
speaker who missed a clause-boundary filled pause and interrupted the
sentence started by the non-native speaker.
They were speaking about a
(14) B-l F-IndonesiaF-USA
Whatever they want it to be they use it for.
NS. It looks like a soccer field though,
It looks more soccerish.
On Monday they will the the they people no my
The students from the ELI play soccer here.
This example showed how disruptive it could be if a planning marker were
not understood by the listener.
In line 4, the nonnative speaker used a
filled pause as a planning marker, but also as a turn-keeping marker.
repetition of her previous utterance.
It took the nonnative speaker two
turns to collect her thoughts and to say the correct words.
recognize them as used by the nonnative speakers.
A possible result was
miscomprehension or miscommunication.
In the following example, the nonnative speaker did not realize that
the "urnm" in line 6 was apparently a planning hesitation and that the native
normative speaker interrupted
encouraging back channel
that showed interest in the subject,
allowed the native speaker to continue despite being interrupted.
(15) H-l F-Taiwan/F-USA
I never saw I never seen it.
NS. It was really good.
NN. Interesting or or excited or what.
NN. Is very very excited or very
Well it it's both. It's
Both? Oh good.
exciting and and it's
it's about love also.
unaware of the turn-holding role of the native speaker's filled pause.
the other hand, native control of language should include an awareness of
illustrate the benefit (example 13) of recognition of this particular function
of filled pauses,
and also the potential miscommunication (examples 14,
that could occur without awareness of this conversational tool.
When switching topics,
and nonnative speakers use
planning filled pauses,
apparently to create additional time to ensure a
proper initiation of topic.
(1980) work with the Pear Stories showed
that when a speaker had to provide new information or describe a change
in focus of a film, the speaker spent more time hesitating.
Additionally, a sudden change in topic could cause the listener to be
In mn fin+ 1n, tH 1c w wtn r nfl^-i nn I\ r, e
onnolrnurn n On 17 d*;f rl+L1 On rl ;n c! ort
illustrated this use of the filled pause.
And that was it right. Ok so um.
What did you.
what you did on Saturday.
NN. I had
That was the da
On Saturday was the day your dad
In this example,
the native speaker changed the subject of the conversation
talking about studying
recent past events
by asking a
By starting with a filled pause,
the native speaker gave both
herself and her listener time to adjust to the new topic.
In this case
filled pause made a transition that was hopefully easier to follow.
A nonnative speaker using a filled pause at a clause-boundary to
initiate a topic switch occurred in example (17).
This was the end of a
discussion about taking the TOEFL.
(17) C-1 F-Taiwan/F-USA
NS. Well, you've taken it once, you know exactly what
r_ nrv V. ^ n An |4 fl flt w? wrnIr rlr nyt yatm nI t r -.nrk^ at I
Hm. Um. Ha- do you know um how to see the um palm
Do I know how to do it?
After several filled pauses and a false start, the nonnative speaker initiated
a new topic, and the native speaker went naturally into the new topic with
in example (18) followed
;. Um Yeah. The um political science is um. For the courses there
is a lot of writing and a lot of term papers um arguing what I'd like
to do really is um,
I'd like to study some like foreign relations
international relations and stuff like that.
mainly what I
wanna focus on.
In this exa
, how long have you been here in Gainesville?
In Gainesville? Um. About like a year and a half.
Year and a half.
Lmple, the native speaker was answering a question about his
studies at the university. In line 6, the nonnative speaker used a filled
pause in to introduce a topic switch, to which the native speaker responded
with another filled pause in line 7, providing even more time to think about
The sneakers in these examples used filled pauses as easy transitions
speakers to plan their thoughts,
but they also provided the listeners with
cues to prepare themselves to accept a topic change.
But on the other side
of the coin, if a speaker quickly switched the topic of conversation without
the other speaker frequently needed to audibly hesitate in
order to plan a response.
An example of this occurred in conversation P-1.
(19) P-l M-Korea/F-USA
Not until then.
I'm gonna go to I'm gonna go play
basketball by myself.
How are your roommates?
Uh my roommate my roommates are pretty good.
After a discussion about playing basketball, the native speaker abruptly
switched the topic with her question in line 5.
The nonnative speaker's
filled pause in line 6 was followed by a repetition of the subject, suggesting
that he needed even more time to recover from the abrupt change and to
plan his answer.
In the following example,
the nonnative speaker suddenly switched
and in response,
the native speaker used several filled pauses,
signalling that she was planning her utterance.
(20) E-1 F-Brazil/F-USA
very pleased. ((laughter)) That is is called self actualizing.
And when you self actualize then you become what you want to be.
Because then you accept yourself fully and you become a whole
person. So, you're doing well.
You're very good.
What kind of film do you like films eh movie do you like?
I I've seen all kinds.
Um of movies.
I don't like the
real horror movies.
NN. I don't either.
I don't like those.
completely unrelated to the topic brought up by the nonnative speaker in
As a result, the native speaker needed to use filled pauses in her
comprehension of the topic and her planning of what to say in response.
In this next example,
while one native speaker was explaining her
language learning experience, the other native speaker changed the topic
in line 3, and the nonnative speaker needed to use a filled pause to create
a little time to think of her answer.
(21) L-1 F-Spain/F-USA
I I took two years of Spanish. I would hope that some of it
Yeah. How did you like Madrid?
I enjoy a lot, you know? I I
But but I was in a (boarding school) in a
Uh huh. Boarding school? No,
it's um boarding school.
both speakers and listeners had strategies for finding time to
When a speaker needed time to initiate a topic switch,
he or she
often used a filled pause.
Similarly, when a listener needed time to think
of information to respond to a topic switch,
she or he often began the next
utterance with a filled pause, buying time for planning a response.
Side Secuences and Interruptions
In addition to planning filled pauses existing as turn-keepers and
with topic shifts,
they also occurred after topic interruptions.
a very loud motorcycle passed by, after which the native speaker
made a sarcastic comment and then used a filled pause before returning to
The filled pause seemed to be a signal that the conversation
would continue as soon as she found the correct place in her thoughts.
(22) R-1 F-Korean/F-USA
NS. You have to have when you're an architecture
student? I don't know that much about it? But I
know people who were in it in the School of
Architecture? Or that's
their major ((loud noise))
Nice motorcycle! Um. But they like they draw um like
sketches for layouts for buildings and stuff like when
you like where the door is and where the bathroom is
Interruptions occasionally came from other speakers interrupting the
conversation, as in the following example.
The two speakers were talking
to each other on the telephone, and the roommate of the nonnative speaker
interrupted the conversation.
(23) H-1 F-Taiwan/F-USA
Hold on please. Hold on.
NN. ((to roommate)) Oh,
, ok. No no no. ((to native speaking
partner)) I get talk eh talking with my roommate.
Hello you hang up? Wake up wake up wake up
In example (23), the native speaker responded to the interruption with an
This was not interpreted as a sufficient response by the nonnative
who tried to revive the conversation by other means.
speaker laughter in the final line was not a joyous laugh,
but more of a
tolerance laugh, such as can be heard after someone tells a bad joke.
suggested that the native speaker became frustrated or possibly irritated
because the native speaker's attempt to return to the conversation
only a filled pause was not accepted by the nonnative speaker.
the speakers recovered from a
int-r y nti-inn
'lhi rmnll ii hp rnmnarpd tn .lTffnrgnn'g (1 (Q7b "'tqireP eniinre* "
speakers had been speaking of churches, then they began to speak about
the tape recording itself. After the nonnative speaker realized that the tape
recorder was on,
to a normal
conversation, the native speaker began with a filled pause (line 5).
(24) H-1 F-Taiwan/F-USA
No, I am recording.
Yeh. Now you are recording?
Um, ok. I'll ask you a question.
interruptions by side sequences followed by audible planning pauses.
the speakers were trying to agree on who a particular person
was in order for the native speaker to continue with her narration of a
(25) P-l M-Korea/F-USA
NS. Yeah. I think. He did wear glasses.
best friend is.
The Car- his
Where where this guy live?
'StP yr t r. a r iI *
courses at the ELI when the native speaker's
returning to the conversation in line
, the native speaker began with a
then a question to clarify the topic.
(26) L-1 F-Spain/F-USA
NN. Ey only
ey for learn English. Grammar
Not so boring. ((laughter))
((interruption from roommate))
NS. Um. So it's
just mainly English courses.
Following an interruption, a filled pause started the utterance
that continued the conversation.
The following example was of a nonnative
speaker responding to an interrupting sneeze by the native speaker.
(27) Q-1 M-SpainrF-USA
NN. Because in y'know I in the EL ELI and in the ELI you speak
with Korean people and sometimes I speak Spanish because there
are some South American people
The conversation was interrupted by the sneeze and the politeness phrases
associated with it.
Upon returning, the normative speaker used the clause-
boundary filled pause before continuing with the topic.
In addition to planning sentences and clauses,
speakers needed to hesitate within sentences. In
native and nonnative
the data collected for
examination, there were fewer native-speaker planning filled pauses than
This suggested, not surprisingly, that native speakers had better
global planning skills than the nonnative speakers.
In some instances of internal planning, it was not uncommon to see
an included question that indicated that the planning was occurring at the
time of the utterance.
In the following example, the native speaker could
previous week, and he combined filled pauses with the question,
it?" to give him time to think of the name.
(28) 0-1 M-Taiwan/M-USA
So, what else is goin on? Have you eaten at the uh the uh
what is it the Palace? The Imperial Boat.
Ah, Imperial Boat.
T~~~~1 ii fb'SI t* 1 1 S rl fl lS a
T II F 11 1 t
planning time generated by the filled pause, she was unable to complete
(29) P-l M-Korea/M-USA
NS. I wanna
see any movies, this week?
see um that movie, I can't remember what
it's called. Oh now I can't remember its name.
On the other hand
, in the next example,
the native speaker was able
to recall the needed word after a brief filled pause.
(30) C-1 F-Taiwan/F-USA
l also l
, I studies uh another city.
Yeah. So I um ((unclear)) I also lived outside.
NS. You uh. Commute. Did you commute?
NS. Drive back and forth to school every day
The falling intonation and brief pause following this native speaker's
pause suggested a rifling though her lexical files to find the correct word.
After finding it,
she blurted it out, then constructed the complete sentence
around the new word.
mi~.- -~---4. r .. -r~.
1~ f ._ ~. I
(31) R-1 F-Kore/F-USA
Your uh your major is uh physical method?
Yeah, I changed it.
I changed my major.
teacher. Maybe. I dunno.
I'm gonna be a
The second filled pause in line
1 showed that the nonnative speaker was
groping for the name of the native speaker's
Apparently they had
spoken of this previously; however
the nonnative speaker forgot exactly
what it was, and she tried to get a clarification of the information.
classroom buildings each speaker knew.
The native speaker was surprised
that the nonnative speaker knew many different ones.
(32) F-l M-Turkish/M-USA
I know. But
see I don't I mean. Did you ever have a class in
See I don't know where the buildings are if I haven't had a class
in them, generally, and
I just ((unclear))
I mean I been to
NS. Little Hall.
West Library and uh,
I don't know Little Hall
. Maybe I've
NS. Little Hall.
NN. but I don't know the name.
Although the nonnative speaker was familiar with the buildings,
had a little trouble remembering the names.
In this case
, the filled pause
gave him time to think of the correct name before saying one in particular.
The speakers in the following example were discussing the problems
of a foreign student getting a driver's license and driving in the
(33) J-1 F-Japanese/F-USA
You wanna drive?
Yeah. ((unclear)) It's
kinda hard for me because I I I sometimes
scare um there's
a (difference) the the the side.
Different sides of the street?
Uh keep uh right side right? In United States.
moments to clarify that she was saying the correct thing.
, rather than planning a single word, the speakers needed
time to think of
a list of related items.
In these cases
both native and
nonnative speakers tended to use filled pauses before each word in the list.
In examples (34) and (35), the nonnative speakers used filled pauses in this
Both speakers were trying to think of a list of items, and each
item was punctuated by an "uh."
(34) N-1 F-Taiwan/F-USA
Somebody introduce music.
Somebody introduce uh their traditional game.
Wha- what col- countries was this? Like.
NN. Um in my case,
uh Puerto Rico, and
we had uh come from ((unclear)) and
uh Spain, and uh Korea, and uh
Taiwan, Japan, and uh Lithuania.
Lithuania and uh Turkey.
a lot of different people.
(35) B-1 F-IndonesiaF-USA
Yeah I should study ((laughter))
You have to study grammar. Like what kinds of words
Uh the passive voice, passive.
Passive voice, uh, conditional if, and relative
I hated grammar. I like vocabulary and er like
spelling and stuff,
NN. Oh, yeah?
In these examples,
the nonnative speakers exhibited a regular pattern of
planning filled pauses in clause-internal positions to give time to plan lists
Note also that in example (35), the native speaker used the same
strategy in line 8.
What this native speaker does closely resembled several
information, and she needed to stop and think about the individual words
-. U_ -1~. U --
T T_ t _- ___ _l -__- t_ _.. _t _. _A.
she had filed away this information as disagreeable,
possibly adding to her
slow recovery of the words.
In the following example, the nonnative speaker was unsure of the
in a filled
(36) N-l F-Taiwan/F-USA
Where do you want to move
Do you know?
I'm not decided um. I just move my um my packing.
4 NN. Uh huh.
The nonnative speaker appeared to be unsure about a word, but was willing
to try to say what she meant,
and there was a hesitation to show that there
The native speaker was aware of that uncertainty, and she
in an appropriate
The filled pause in this situation might be signalling to the
Doughty (1987) reported that this type of interactional negotiation between
speakers was an important aid in comprehension for nonnative speakers.
In example (37), the nonnative speaker searched for the word "meal"
His conversation partner did not correct him; apparently she
fully understand theft manin r
the aennd native snenker.
chose to provide the word,
nonnative speaker responded positively.
(37) Q-1 M-Spain/F-USA
So big. It's
the biggest uh food in the day.
Um is the biggest meal in the day
. So you
spend two hours or more.
speaker uttered the entire sentence over again,
boundary filled pause, using the correction.
(38) C-1 F-Taiwan/F-USA
speaker showed a similar
Jh. Makes you homesick?
Yeah! I I could I couldn't find uh uh in-
ingredients in here. Some some ingredients. Only a
few. So uh
NS. Can't find all the ingredients you can get at home.
I'm I really really want go
home, but my mother says you stay there. It's ok.
This sensitivity to the needs of the nonnative speaker was frequently seen
nonnative speakers, the native speakers were helping them out when they
Example (38) showed the native speaker doing just that.
It was shown by some researchers that help of this kind between two native
speakers was also a signal of solidarity (e.g. Boxer
In the following example, however, the non
native speaker did not
know the word he was looking for, and the native speaker was unable to
(39) 0-1 M-Taiwan/M-USA
Tomorrow, tomorrow there is a good buffet.
we should go.
M uh I I don't know in English how to call it's
6 II just know Chinese name.
7 NS. Oh yeah?
In this case, the native speaker did not have enough of a clue to help the
nonnative speaker by providing the new word.
continued to give the Chinese name,
was of no help to the native
Approximately one minute following this brief excerpt was spent
unsuccessfully trying to negotiate a meaning
for this word.
r/n n .-an1 nyltjm n .' A nion n i ni onn1n f110A noion +b of 0k Q+' O
were a number of
examples of filled pauses used in this manner.
Examples (40) and (41) were typical examples of repairs.
(40) B-l F-Indonesiag-USA
Where where uh when is your last class
My last exam you mean?
My no. In two weeks we will finish.
I have classes all through through dead
5 week I have class.
(41) H-1 F-Taiwan/F-USA
Yes, but uh we have we uh have one free free the
The TOEFL test.
understand my meaning.
NS. See I understand, yes.
NN. We have uh uh we will be have on the April 24th.
NS. April 24th.
The nonnative speaker in example (40) started the sentence twice, before
hesitating to reconfigure her sentence,
using the filled pause before the
repair, as did the nonnative speaker in example (41).
In both cases,
appeared that the native speakers easily understood the repairs, suggesting
that this strategy worked well.
strategy for repairing errors, a
Some native speakers also used the same
s in example (42).
(42) D-1 M-Panama/M-USA
little English before you got here or
In this case
all I know already.
, the native speaker added the phrase
"I mean" to his filled
pause to insure that the nonnative speaker recognized that there was a
corrected utterance to follow.
speech without signalling in any way that they had made changes.
Example (43) showed a native speaker correction with no filled
pause as a signal.
(43) F-1 M-Turkey/M-USA
But in the United States, it's
not the same.
not the same.
That's what my a Turkish guy he's
a computer engineer?
And he's gonna go back to Turkey and open up his
computer business. A computer business.
His own business.
In line 4, the native speaker made a correction, but he did it very quickly
with no hesitation at all.
There was some evidence that the nonnative
clarification comment in line
. Often these unmarked corrections by the
na-170 .nglwrvc wro rnniyrl nndl in th middrllns a f n lanorr iitrPrnfnr1 Indr f.hn