ChapterPDF Available

Discourse as an interactional achievement: Some uses of `uh huh' and other things that come between sentences

Authors:
IN
Georgetown
Universit
y
Roundtable
on
Languages
and
Linguistics
1981;
Analyzin
g
Discourse
Text
and
Talk
Ed
.
by
D
.
Tannen
Georgerown
Uni
-
veraity
Press,
Washington
D
.C.,
1982
DISCOURSE
AS
AN
INTERACTIONAL
ACHIEVEMENT
:
SOMEUSES
OF
'UH
HUH'
AND
OTHER
THING
S
THAT
COME
BETWEEN
SENTENCES
Emanuel
A
.
Scheglof
t
University
of
California
.
Los
Angeles
1
.
From
the
standpointof
students
of
'discourse',
conver-
sation
and
other
forms
of
talk in
interaction
are
subvarietles
of
discourse
.
Whatcomes
to
be
minimally
criterial
for
discourse
is
the
presence
.in
some
sort
of
coherent
relationship,
of
a
spate
of
language
use
composed
ofmore
than
one
sentence
(or
whatever
other
unit
is
treated
as
grammatically
fundamental)
.
It is
then
common
to
be
concerned
(1)
with
the
basis
for
the
apparent
coherence
between
the
several
components
of
the
dis-
course,
(2)
with
the
cognitive
structure
of
the
unit,
and
(3)
with
the
mechanisms
by which
it
in
analyzedor
decoded
on
reception
.
There
may
be an
effort to
discern quasi-syntactic
relationships
between
successiveparts
of
the
discourse
-
-be-
tween
successive
sentences,
for
example
.
And
discourse
units,
such
as
paragraphs,
may
be
found
to
be
constituted
by such
quasi
-syntactic
relationships
.
The
actual
enactment
of
the
discourse---for
example,
its
telling
--often
seems
to
be
treated
as
the
behavioral
realization
of
a
preplanned
cognitive
unit.
The
prototypediscourse
for
such
an
approach
is
the
narra-
tive
or
lecture,
which
readily
lends
itself
to
treatment
as
the
product
of
a
single
speaker,
whose
cognitive
apparatus
under-
lies
and
shapes
it
.
For
the
student
of
talk in
interaction
.
dlscoárae
(still
mini-
mally
defined
as a
spate
of
talk
composed
of
more
than
one
sentence
or
other
fundamental
unit)
in
more
usefully
treated
as
one
type of
production
in conversation
(or
other
speech-
exchange
situation)
.
Note
that,
although
some
sorts
of
ob-
jects
for
analysis,
such
as
written
stories,
memoranda,
and
legal
documents,
may
appear
suitable
for
analysis
under
the
former conception
but
not
the
latter
.i
n
many
cases
the
same
objects
of
inquiry
are
seen
d1ffernt1y
from
the
two
points
of
view
.
Thecommon
discourse-analytic standpoint
treats
th
e
71
Schegloff's homepage
Click for sound excerpts
72
I
Emanuel
A
.
Schegloff
lecture
.or
sermon
.
or
story
told
in
an
elicitation
interview,
camptre
setting,
or
around
the
table,
as
the
product
of a
single
speaker
and
a
single
mind
;
the
conversation
-analytic
angle
of
inquiry
does
not
let
go of the
fact
that
speech
exchange
systems
are involved,
in
which
more
than
one
par-
ticipant
is
present
and
relevant
to
the
talk,
even
when
only
one
does
any
talking
.
Let
me
recount
an
old
experience
.
Once
I
had
trouble
understanding
certain
monologues
in
Shakespeare
.
I
was
watching
a
series
of
rehearsals
of
The
Winter's
Tale
by
the
Canadian
Shakespeare
Company
on
public
television,
and
had
gotten
down
my
Complete
Shakespeare
and
was
following
the
text
. In the
monologue
in
question.
I
could see
how
line
2
followed
1,
3
followed
2,
and
line
4
followed
3
;
but
I
just
could
not
figure
out
how
line
5
followed
line
4.
And
then
I
saw
in
a
series
of
rehearsals
that,
the
authoritative
text in
front
of
me
to
the
contrary
notwithstanding
.
line
5
did not
follow
line
4;
some
action
followed
line 4,
and
line
5
followed
that action
.
And
what
was
at
issue
in
the
rehearsals
was
what
that
action
should
be
and
who
should
do
it
.
for
the
sense
of
line
5,
and
ensuing
lines,
would
be
affected
in
a
major
way
by
it
.
Anyone
who
has
lectured
to
a
class
knows
that
the
(often
silent)
reactions
of the audience
-
-the
wrinkling
of
brows
at
some
point
in
its
course,
a
few
smiles
or
chuckles
or
nods,
or
their
absence
-can
have
marked
consequences
for
the
talk
which
follows
;
whether,
for
example,
the
just
preceding
point
is
reviewed,
elaborated,
put
more
simply,
ate
., or
whether
the
-
talk
moves
quickly
on
to
the
next
point,
and
per-
haps
to
a
more
subtle point
than
was
previously
planned
.
1
If
this
is
the
case
in
such
a
situation
of
talk
-in
-interaction
as
the
lecture,
then
its
relevanceshould
be entertained
as well
for
experiments,
elicitation
interviews,
and
ordinary
conver-
sation
.
Clearly,
different
speech
-exchange
systems
are
involved
in
lectures
and
ordinary
conversation
w
ith
different
turn
-takin
g
practices
providing
quite
differently
structured
opportunities
to
talk
or
participate
in
other
ways
.
That
is
one
reason
why
the
reference
to
lecture
situations
describes
wrinkled
brows
and
smiles
and
nods,
rather
than
utterances or
even
'uh
huh's
.
Clearly,
as
well,
in
several
different
types
of
speech
-exchange
situations
.there
can
be
occasions
in
which
participation
is
constructed
by
a
speaker
in
continuing
response
to
inter-
actional
contingencies
and
opportunities
from
moment
to
mo-
ment
.
and
occasions
in
which
a
participant
has
a
preformed
notion,
and
sometimes
a
prespeciLed
text,
of
what
is to
be
said,
and
plows
ahead
with
it
in
substantial
(though
rarely
total)
disregard
for
what
is
transpiring
in
the
course
of
this
talking
.
But
these
two
extremes
are not
equally
likely
to
occur
in
the various
types
of
speech
-exchange
situations
;
the
prespecitled
text,
adhered
to
'no
matter
what',
is
much
less
Discourse
as
an
Interactional
Achievement
I
73
common
(and
for
good
structural
reasons)
in ordinary
conver-
sation
than
in
sermons
or
lectures
.
Even
the
wholly
prespeci-
fled
al,
which
most approximates
the
enactment
of a
cognitive
object,
must
be
adapted
in
its
delivery
to
its
occasion,
and
will
certainly
have been
designed
with
attention
to its
recipients
and
the
situation
of
its
delivery
in
the
first
place-
-both
aspects
of
Interactional
sensitivity
.
However
.
It
should
be
clear
at
the
outset
that
in
what
follows
I
am
most
centrally
concerned
with
what
I
take
to
be
both the
primordial
and
the
most
common
setting
and
organization
for
the
use
of
language
-
-ordinary
con-
versation
.
Although
much
of
what
I
have
to
say
is
relevant
to
other
settings,
the
way
in
which
orientation
to
co
-participants
and
Interactlonal
structure matter
to
discourse
and
its
forma-
tion
.
will
vary
in
different
speech
exchange
systems
with
differ-
ent turn
-taking
systems
.
Important
analytic
leverage
can
be
gained
if
the
examination
of
any
discourse
is
conducted
in
a
manner
guided
by
the
following
.
2
U)
The
discourse
should
be treated
as
an
achieve-
ment
;
that involves
treating
the
discourse
as
something
'pro-
duced'
over
time,
incrementally
accomplished
.rather
than
born
naturally
whole
out of the speaker's
forehead,
the
delivery
o
f
a
cognitive
plan
.
(2)
The
accomplishment
or
achievement
i
s
an
Intaractional
one.
Quite
aside
from
whatever
individual
cognitive
or
processing
achievements
might
be
involved
(which
are not
to
be
treated
only as
anterior
to
the
Interactional),
the
production
of a spate of
talk
by
one
speaker
is
something
which
involves
collaboration
with
the
other
parties
present,
and
that
collaboration
is
interactive
in
character,
and
inter-
laced
throughout
the
discourse,
that
in
.i
t
is
an
ongoing
ac-
complishment,
rather
than
apact
signed
at
the
beginning
.
after
which
the
discourse
is
produced
entirely
as
a matter of
individual
effort
.
(3)
The
character
of
this
Interactional
ac-
complishment
is at
least
in part
shaped
by
the
sociosequentia
l
organization
of
participation
in
conversation,
for
example
by
its
turn
-taking
organization,
which
in
not
organized
to
be
in-
different
to
the
size
of
the
turnS
parties
take, but
whose
underlying
(though
supercessable)
organization
is
designed
to
minimize
turn
size
.
It is
this
feature
which
requires
us
to
see
'discourse'
and
'discourse
units'
which
have overcome
this
bias
as
achievements
and
accomplishments
.
(4)
Because
the
actual
outcome
will
have
been
achieved
by
the
parties
in real
time
and
as,
at
each
point,
acontingent
accomplishment,
the
mechanisms
of
the
achievement
and
its
effort
are
displayed,
or
are analyzably
hidden
in
or absent
from,
various
bits
of
behavior
composing
and
accompanying
that
discourse,
and
analyzable
with
it
.
One
class
of
such behavior
which
is
implicated
in
the
achievement
of
discourses
in
conversation
is
the
concern
of
this
paper
.
Instances
of
the
class
take
the
form of
vocaliza-tion
such
as
'uh huh',
'mm
hmsn',
'yeah!,
and
others
as
well
74
/
Emanuel
A
.
Schegloff
is
head
-gestures
such
as
nods
.'
These,
as well
as
other
.
sits
of
talk
and
behavior
produced
by
other
than
the 'main
speaker'
are
regularly
discarded
when
discourses
the
stories,
the
arguments,
etc
.
--are extracted
from
the
tangle
of
detail
which
composed
their
actual
occurrence. 'The
story'
is
pun-
ed
of
them
in
the
course
of
its
extraction,
both
by
lay
re-
counters
andby
professional
analysts
.
It in
this
separation
of
bits
of
talk,
otherwise
intercalated
with
each
other
and
con-
tingent
on
one
another
.
Into
two
distinct
classes,
of
which
one
is
the
'real talk'
(the
story,
the 'what-was
-being
-said')
and
the
other
conversational
'detritus'
(apparently Lacking semantic
content
.
and
seemingly
not
contributing
to
the
substance
of
what
the
discourse
ends
up
having
said)
.
which
makes
possi-
ble
the
notion
of
'discourse'
as
a
single
speaker's,
and
a
single
mind's,
product
.
It
is
a
consequence
as
well
that
the inter-
actional
animus
and
dynamic
of the
spate
of
talk
can
disappear
into
the
cognitive
structure
and
quasi
-syntactic
composition of
the
discourse
.
What
has
been
discarded
may
itself
be
picke
d
up
by
investigators
-
-typically
other
investigators,
even
othe
r
sorts
of
investigators
--
for
separate treatment
under
such
rubrics
as
'accompaniment
signals'
(Kendon
1967)
or
'back-
channel'
actions
(Yngve
1970
;
cf
.
Duncanand
Flake
1977)
.
But,
as
I
urge
later,
the
fact
that
both parts
of
the occa-
sion-the
teller's
telling
and
the behavior of the
recipients
-
maymay
be
subjected
to
study
does
not
restore
the
interactivit
y
lost
when
the
former
to
extracted from the
latter
.
For
th
e
parts of the
telling
appear
to
follow
one
another, rather
tha
n
each
following
some
responsive
behavior
by
a
recipient
(or the
lack
thereof)
;
and
what
recipients
produced
after
this
or
that
part
of the
telling
has
been
removed
from the
environment
o
f
that
to
which
it
was
responsive
.
From
'discourse'
and
'listene
r
behavior'
so
conceived
and
studied,
It
in
unlikely
that
one
wil
l
be
able
to
reassemble
the
actual
structure of
'talking
at
Lengt
h
in
conversation'
.
In
what
follows,
I
first
elaborate
a
bit
on
the
meaning
of,
and
the
reasons
for,
treating
the
occurrence
of
discourse
in
conversation as 'an
achievement'
.
One
mechanism
for
that
achievement
has
its
ocus
at
points
at
which
recipients
or
hearers
could
begin
talking
but
content
themselves
with
'uh
huh'
and
the
like
Instead,
after
which
prior
speakers
continue
.
I
briefly
discuss
recent
treatments
of
vocalizations
such
as
'uh
huh'
and
'yeah',
and
then
offer
some
alternatives
.
2
.
Why
should
the existence of a
'discourse'
(a multi-
sentence
unit)
in
ordinary
conversation
be
treated
as
an
achievement?
Elsewhere
(Sacks,
Schegloff,
and
Jefferson
1974)
it
has
been argued
that
speakers
construct utterances
in
turns
at
talk
out
of
describable structured
units,
with
recognizable
possible
completions
.
In
English,
some
lexical
items
(e
.g.
'hello',
'yes',
'who'),
some
phrasal
units,
some
clausal
units,
and
sentences
constitute
such
'turn
-constructiontil
Discourse
as
an
Interactional
Achievement
/
75
units'
.
The
end
of
any
such
unit
in
a
possible
completion of
a
turn
.
and
possible
completions
of
turns
are
places
at
which
potential
next speakers
appropriately
start
next
turns
.
If
this
in
the
case, then an
underlying
structure
of
turn
distri-
bution
is
in
operation
which
organizes
interactive
enforcement
.
or
potential
enforcement,
of
a
minimization
of
turn
size
.
If
such
a
system
is
in
operation,
then
a
constraint
for
single-unit
turns
has
at
least
two
sources
.
First,
there
in
an
organizational basis
for
current
nonspeakers
to
monitor
for
the
possible
completion
of
first
units
in
a
current
turn
as
a
place
to
start
next turns
.
And
second,
there
in
an
orientation
by
speakers
starting
a
turn
to
the
organizationally
motivated
orien-
tation
of others
to
so
start
up,
which
can
engender
a
design-
ing
of
the
talk
in a
turn
to
be
so
organized
as
to
get
what
needs
to
be
said
said
before the
end
of
the
first
unit's
com-
pletion
.
The
second
of
these
can
contribute
to
making
the
smooth
operation of
the
first
viable
and
routine
.
A
great
many
turns
at talk in
conversation
thus
end
up
being
one
unit long
.
With
all
of
this,
it
in
obvious
that
some
turns
at
talk
end
up
having
more
than
one
unit in
them
.
Nor
in
this
an
anomaly,
or
counterevidence
.
It
does
invite
exploration
of
the
possible
existence
(and
features)
of
methodical
ways
in
which
such
multi-unit
turns
are
achieved.
If
there
are
such
ways,
then
their
use
may
serve
as
additional
evidence
of
the
underlying
organization
which
these
methods
are
used
to
supersede,
at
the
same
time
as
they
explicate
the
work
of
achieving
the
aupersession
-
-the
discourse
.
Although
this in
not the
place
to
undertake
an
extensive,
let
alone
exhaustive,
account
of
such
methodical
devices, several
may
be
mentioned
to
suppl
y
"
sense
of the sorts of
phenomena
that
are
involved.
One
class
of
methods
by
which
multi-unit
turns
may
be
achieved
is
that
composed
of devices
initiated
by,
the
potential
speaker
of
the
multi-unit
turn.
(a)
The
potential
discourse
-speaker
may
indicate
from
the
beginning
of
the
turn
an
interest
in
producing
a
more
-than-
onà
-unlt turn
.
For
example,
the
speaker
may
begin
with a
list
-initiating
marker,
such
as
'first
of
all','
projecting
thereby
that
after
the turn-unit
in
which
the
'first'
is
done,
more
wili
follow
.
Note
that
there
may
otherwise
be no
particular
need
to
pre
-mark
an
item
as
first
in
a
list
(I .e
.
besides
leaving
it
to
be
so
discovered
over
the
course
of
extended
talk,
by
vir-
tue of
eventual
subsequent
items)
other
than
the
problem
of
getting
to
produce
subsequent
items
.
Beginning
a
turn
this
way
recognizes
the
turn
-taking
contingency,
and,
by
project-
ing
a
multi-unit
turn,
invites
recipients
to
hold
off
talking
where
they
might
otherwise
start,
so that
the
'post
-first
-units'
may
have room
to
be
produced
.
(b)
Indeed,
the
turn
-position,
or
turn
-opportunity,
in
which
the
beginning
of a
projected
multi-unit
turn could
be pro-
duced,
may
instead
be
entirety
devoted
to
a
whole
turn
which
is
focused
on
doing
the
projecting
(as
the
list -Initiating
marker
76
I
Emanuel
A
.Schegloif
does
within
a turn)
.
Some
years
ago,
Sacks
(1974)
describe
d
one
such
operation
under
the
rubric
'story
prefaces',
and
more
recently
.
I
have
described
under
the term
'pre
-pre'
a
similar
logic
of
use
underlying
one
class
of
occurrences
of utterances
of the form 'Can
I
ask
you
a
question'
(Schegloff
1980)
.
In
both
cases,
a
course
of
talk
is
projected
which
involves
more
than
one
turn
-constructional
unit,
and
the
talk
begins
with a
display of
that
projection
.
Note
that
it
remains
for
recipients
to
honor
this
projection,
and
to
withhold
talk
at
the
points
at
which
it
would
otherwise
be
appropriate
.
Although
initiated
by
the
Intending
extended
-turn
speaker
.
If
an extended
turn
results,
it
will
have
involved
interactive
accomplishment
by
both
speaker
and
recipients,
the
latter
being
recipients
only
by
abjuring
their
possible
status
as speakers.
The
list-
initiating
marker,
or story preface, or
'pre
-pre'
(I .e
.
'Can
I
ask
you
a
question')
are the
overt
markers
of
orientation
to
the
constraints
making
achievement
of discourse
problematic
.
and
of the
effort
directed
to
superseding
them
.
(c)
Speakers
may
also
employ
methodical
devices
for
achiev-
ing
a
multi
-unit
turn
at
positions
other
than the
beginning
of
the
turn
in
question
.
There
is, for
example,
what
can
be
called
the
'rush
through'
-
-a
practice
in
which
a
speaker,
ap-
proaching
a
possible
completion
of
a
turn
-constructional
unit,
speeds
up
the
pace
of
the
talk,
withholds a
dropping
pitch
or the
intake
of
breath,
and
phrases
the
talk
to
bridge
what
would
otherwise
be
the
juncture
at
the
end
of a
unit
.
In-
stead,
the
speaker
'rushes
through' the juncture without
in-
breath, reaches
a
point
well Into
a
next
unit
(e
.g
.
next
sen-
tence),
and
there
stops
for
:a
bit,
for
an
inbreath,
etc
.
(Schegloff
1973)
.
Here
the
turn
-extension
device
is
initiated
near
the
otherwise
-possible
-end of the
turn, rather
than
at
its
beginning
.
Once
again, interaction
and
collaboration
are
involved,
for
recipients
could
start
up
despite
the
displayed
intention
of
current
speaker
to
continue,
and
produce
thereby
at
least
a
'floor light'
.
Once
again,
the
turn
-extension
de-
vice exhibits,
on
the speaker's
part,
an
orientation
to the
imminent
possibility
of
another
starting
up
as
s/he
approaches
the
end
of the
turn
-unit.
Once
again,
it
successful
at
getting
to
produce
a
multi-unit
turn
or
discourse,
the
talk
displays
the
special
effort
involved
in
achieving
it
.
Of
course, not
aU
mufti
-unit
turns
are the
result
of
speaker-
initiated
methods
designed
to
achieve
them.
Some
multi-unit
turns
are
the
outcome
of a
different
methodical
production
.
A
speaker
produces
a
one
-unit
turn,
at
whose
possible
comple-
tion
no co
-participant
starts
a
next
turn.
Then
one
way
the
talk
may
continue
is
by
the
prior
speaker
talking
again,
some-
times
by
starting
a
new
turn
unit
.
On
possible
completion
of
the
now
added
second
unit,
a
multi-unit
turn
has
been pro-
duced
;
of
course,
the
same
cycle
may
occur
on
the next
possible
completion
as
well
.' In
cases
of
this
sort,
the
course
of
action
which
issues
in
a
multi-unit
turn
is
'initiated'
by
a
4
Discourse
as
an
Interactional
Achievement
/
77
recipient,
and
not
by
an
(intending)
speaker
of
(what
ends
up
as)
the
multi-unit
turn,
or
discourse
unit
.
Once
again,
interactions!
achievement
is
involved,
each
participant
orient-
ing
to
the
other(s),
and
all
oriented
to
the
underlying turn-
taking
organization
which
is itself
an
Interactionaily
driven
and
constrained
organization
.
Once
again,
signs of
the
col-
laborative
work
are
marbled through
the
talk
-
-In
this
case.
In
the
form of a
frequent
slight
gap
of
silence
at
the
possible
turn
completion
which
can
issue
in
prior
speaker
resuming
and
extending
the
turn
into
a
multi
-unit
on
e
In the
preceding
.
I
have
tried
to
point
to
several
methodica
l
routes
by
which
multi
-unit
turns
or
discourses
can
come
to be
.
Each
concerns
how
a
second
turn
-constructional
unit
can
com
e
to
be
produced
at
the
point
at
which
the
underlying
turn
-
taking
organization
otherwise
provides for
turn-transition
.
But
sometimes
quite
extended
spates of
talk
are involved-
-
stories,
chains of
argument,
long
descriptions,
etc
.
Th
e
point
about
the
joint,
interactive
achievement
of
discourse
is
not
limited
to
the
beginning
of
spates of
discourse
-the
the
initia
l
possible
transition
point
at
which
the
turn
stays
with
the
sam
e
speaker
.
Recurrently
through
an
extended
spate
of
talk
,
places
where
others could
start
up
appear,
and
when
other
s
do
not
start
up
full
utterances,
there
are
commonly
smaU
be
-
havioral
tokens
by
which
interactive
management
of
the
possi
-
ble
transition
occasion
is
effected
-bits
of
assessment
or the
absence
of
them
where
they
are
relevant
.tokens
of
interest
,
nods,
smiles
.'uh huh's,
and
withholding of
these,
gaze
direc
-
tion
with
or
without
mutual
gaze,
and
the
like
.
It is
on
on
e
class
of
these
that
I
concentrate
in
what
follows
.
3.
The
modern
literature
in
which
bits
of
talk,
vocalization
or
related
behavior
are
extracted
from
what
becomes
ongoing
talk
by
another,
and
are
subjected
to
treatment
in
the
aggre-
gate,
begins
with the
linguist
Fries (1952)
.
Fries
treated
to-
gether
the
following
sorts of
forms
(1952
:49)
:
'yes'
.
'unh
hunh',
'yeah',
'I
see',
'good'
.
'oh',
and
others
of
lesser
fre-
quency
.
Others have
dealt
with
body
-behavioral
versions
of
this
behavior,
and
have
discussed
the vocalized
forms
in
the
course
of
doing
so
(Kendon
1967
;
Dtttman
and
Llowellyn
1967
.
1968)
.
The
most
common
term
now
in
use
for
such
item
.
'back-channel
communication'
.
was
introduced
by
Yngve
(1970)
.
d
includes
a
much
broader
range
of
utterance
types, In-
Clud .Ing
much
longer
stretches
of
talk
.
The
term
'back-
channel'
has
been
adopted
by
Duncan
and
his
associates
(for
example,
Duncan
and
Fiske
1977),
together
with
the
broad-
ened
definition
of
the
class
.
Duncan
and
Fiske
(201
-202) in-
clude not
only
expressions
such
as
'uh huh',
'yeah',
and
the
like,
but
also
completions
by
a
recipient
of
sentences
begun
by
another,
requests
for
clarification,
'brief
restatement'
of
something
just
said
by
another,
and
'head
nods
and
shakes'.
78
/
Emanuel
A
.
Schegloff
Throughout
this
literature,
two
related
characterizations
have been
offered
to
deal with
these
bits
of
behavior
.
Ac-
cording
to
one,
these
bits
of behavior
are
evidence
of atten-
tion,
interest,,
and/or
understanding
on
the
listener's
part
.
(Thus
Fries
1952
:49,
'..
.signals
of
this
continued
attention
..
.',
or
.
Kendon
1967
:44
.
I-
appears
to
do no
more than
signal
...
that
he
is
attending
and
following
what
in
being
said
...')
.'
A
second
use
of
such
behavior
proposed
in
this
literature
in
that
it
'
.
.
.keeps
the
conversation
going
smoothly'
(Dlttman
and
t
.leweUyn
1967
:342),
or .
.
.
appears
to
provide
the
auditor
with
a
means
for
participating
actively
in
the
conversation,
thus
facilitating
the
general
coordination
of
action
by
both
participants
...'
(Duncan
and
Fiske 1977
:202
-203)
.
1
do not Intend
to
comment
extensively
on
this
second
characterization
beyond
noting
that
once
an
organization
of
conversation
is
established
in
which nonspeaker
Interpoaltiona
are a
recurrent
part,
their
presence
will
be
part
of
'going
smoothly'
or
of
active
participation
;
but
this
does
not
tell
us
why
active
participation
in
taken
to
involve
this sort
of
be-
havior,
or
why
the
absence
of
such
interpolations
undercuts
the
'smoothness'
of the
conversation
.
If
indeed
it
does
(cf
.
Schegloff 1968:1092
-1093)
.
However,
It
is
the
capacity
of
'uh
huh'
and
cognate
bits
of
behavior
to
betoken
attention
and
understanding
which
is
the
most
common
proposal about these
events
taken
in
the
aggregate,
with
each
removed
from
its
context
of
occurrence
;
and
it
is
to
this
sort
of
characteriza-
tion that
the
following
points
are
addressed
.
(1)
The
term often
used
in
the
literature
to
describe 'uh
huh'
and
similar
productions
is
'signal',
and
it
is
unclear
what
the
Implications
of
this
term
are
for
the
strength
of
what
is
believed
to
be
done
by
these
bits
of
behavior
.
It
is
worth
noting,
however,
that
'uh
huh',
'mm
hmm',
'yeah',
head
nods,
and
the
like
at
best
claim
attention
and/or
understanding
.
rather
than
showing
it
or
evidencing
it
.
The
references
to
'signals
of
continued
attention'
or
'signal
..
.
that
he
is
attend-
ing
and
following'
treat
these as
more
than
claims,
but
as
cor-
rect
claims,
and
this
need
not be the
case
;
It
in
.
at
any
rate,
a
contingent
outcome,
and
not an
intrinsic
characteristic
of
the
behavior
being
described
.
(2)
It
is
unclear
why
any
particular
behavior
-
-auch
as 'uh
huh' or a
head
nod-should
should
be needed
to
address
the
issue
of
attention,
whether
to
claim
it
or
to
show
it
.
Regularly,
thes
e
bits
of
behavior
are
produced
when
there are otherwise
pres
-
ent
on
a
continuous
basis sorts
of
behavior
which
are
under
-
stood
as
manifestations
or
exhibits
of
attention,
such
as
con
-
tinuing gaze
direction
at
speaker
.'
Aside, then,
from
the
issue of
whether
'uh
huh'
etc
.
evidence
attention
or
claim
it
,
there
is
the
issue
of
why
attention
is
taken
to
be
problemati
c
in
the
rst
place, in
need
of
showing
or
claiming
.
(3)
If
.
for
the
moment,
we
treat
the
issue
of
attention
as
having
its
relevance
established,
then
it
may
be
noted
that
any
instance
of
an
indeñnitelv
extendable
set
of
utterances
Discourse
as
an
Interactlonal
Achievement
/
79
would
either claim
or
show
attention
to
or
understanding
of,
an
immediately
preceding
utterance
by
another
.
That
is
.
a
vast
array
of
types
of
talk
following
an
utterance
by
another
exhibit
an
orientation
to
it ;
accordingly,
the
claim
that
'uh
huh'
exhibits
an
orientation
to,
or
attention
to,
preceding
talk
does
not
help
discriminate
'uh
huh'
from
any
other
talk,
or
tell
us
what
'uh
huh'
in
particular
does
or
can do,
and
there-
fore
why
a
participant
mightchoose
to
produce
it
rather
than
something
else
.
(d)
If,
however,
we
aim
to
understand
how
bits
of
behavior
such
as 'uh
huh'
and
the
like
may
be
taken
as
bearing
on
the
attention
. Interest,
or understanding
of
their
producers
with
respect
to
the
talk
beingproduced
by
another, then
we
should
also
note
that
'uh
huh',
'yeah',
and
the
like
are regularly
taken
as
betokening
agreement
as
well
. A
search
for
the
mechanism
by which
interest
.
attention,
or
understanding
are
exhibited
by
this
behavior,
should
also
deal
with
the
apparent
exhibiting
of
agreement
.
When
'uh
huh's
etc
.
are
considered
in the
aggregate,then
,
the
characterization
of
the
class as
signaling
attention,
inter
-est,
or
understanding
appears
equivocal
.
Although
it
can
be
argued
that attention
and
understanding
are
generically
rele
-
vant
in
conversation,
no ready
account
in at
hand
(when
th
e
aggregate
of
cases
is
considered)
for
why
these
issues
nee
d
specially
to
be
addressed,
why
they
are
addressed
with
thes
e
tokens,
why
addressed
at
these
particular
points
(If
.
Indeed
.
it
is at
particular
points,
on
this
account,
that
these
token
s
are
placed)
.
However,
examination
of
particular
occurrences
of
the
sort
of behavior
underdiscussion
-of
particular
'uh
huh's,
'yeah's,
etc
.-
-might
yield
answers
to
some
of
these
questions
.i
n
particular
Inst
e
.nces,
for
example,
analysis
may
show
that
attention
was
indeed
problematic
for
the
parties,
and
that
an
'uh
huh' or
a
nod
was
produced
'In
response
to'
anextended
gaze
by
the
speaker
whichappeared
to
solicit
a
sign
of
attentlon/intereat/understandlng
.
Or,
analysis
may
show
that
certain
usages
by
speakers
regularly involve
addressing
the
Issue
of
understanding
in their
immediate
aftermath
.
Thus,
as
describedelsewhere
(Sacks
and
Schegloff
1979)
.
speakers
may
use
'recognitional
reference
forms'
(such
as
propernames)
to
refer
to
persons
they
think
recipients
know
;
but
if
speakers
are not
certain
that
recipients
know
the
intended
referent,
they
may
mark
the
reference
form with
an
upward
intonation,
soliciting
some
signal
of
recognition
(a
special
kind of
under-
standing)
;
if
no such
display
l
forthcoming,
further
tries,
involving
further
clues
to
the
identification
of
the
referent
.
are
provided,
with
display
of
recognition
again
solicited
.
Re-
cipients
may
betoken
such
a
recognition with
'uh
huh' or
may
add
to
this
token
(especially
if
recognition
was
delayed)
Some
demonstrationof
recognition,
as in
(1)
and
(2)
.
80
/
Emanuel
A
.
Schegloff
(1)
A:
Ya
still
in
the
real
estate
business.
Lawrenc
e
B
:
-,
Wab
&
ub
no
my
dear
heart
uh,
ye
knowMax
lUckier
h
-
(0 .5)
B:
with
whom
I've
been
'ssociated
since
I've
been
out
here
in
Brentwood
has
had
a
series
of
urn (0
.
?)
A
:
-'
"
(yeah
B
;
=bad
experiences
uhhhhh
I
guess
he
calls
it
a
nervous
breakdown
.
A
:
Yea
h
(Sacks
and
Schegloff 1979
:19)
(2)
L ;  .
.
.
wefl
I
was
the only
one
other
than than
the
uhn
i
+
tch
Fords
Ub
Missis
Holmes
Ford?
You
know
uh
the
the
cellist
?
W
:
+
Oh
yes
.
She's
she's
the
cellist
.
L
: 
Ye
s
(Sacks
and
Schegloff
1979:19)
With
this
background,
one
can
note
that
even
in the
absence
of overt
Solicitation
ky
upward
intonation
of
some
display
of
recognition,
after
recognitlonal
reference
one
commonly
finds
'ub
huh'
and
the
like,' as
in
(3)
.
(3)
Bee
:
hh
This
War
I
have
-
(Iv
-) "felluh"
this
ma
:n
.
(0.2)t!bhhlieha::(s)-uff-eb-who-whoIhave
fer
Un4tics
is
real ly
too
much,
hh
h
Ava
;
- 
[Mm
bin?'
(Mm
bin
,
(TG
.
190
-201)
It
is
not
that
some
substantial
proportion
of 'uh huh's
etc
.
are
thus
accounted
for,
but
that
an
analytically
coherent
set
of
cases
can
be
assembled
in
this
way
from a
series
of
analyses
of
individual
cases,
the
basis
for
the
coherence
of the
class
being
derived
from
the
sequential
environment
in
which
those
particular
tokens
are
produced
.
Although
appeals
to
signalling
attention,
interest,
and/or
understanding
appear
equivocal
when
invoked
on
behalf
of the
aggregated occurrence
of tokens
such
as
'uh
huh',
'yeah'
.
and
the
like
removed
from
their
particular
environments,
such
accounts
may
be
viable
and
strong
when
introduced
for
delimited
and
described
cases
in
which
the
rele-
vance
of
these
issues
for
the
parties
to
the
conversation
at
that
point
in
the
talk
can
be
shown
.
Appropriate
sets
of
such
analyzed
single
cases
may
then be
assembled
to
display
re-
current
practices,
themes,
structures,
etc
.
l .
Is
there
nothing
more
general,
then,
that
can
be
said
about
such
utterances
as
'ub
huh'
and
the
like,
when
they
compose
all
of
their
producer's
vocalization
on
that occasion
Discourse
as an
Interactiorial
Achievement
I
al
of
talking?
Two
observations
seem
to
me
to
have
sufficiently
general
relevance
to
bear
mention
in this
connection
.
Perhaps
the
most
common
usage
of 'uh huh',
etc
.
(in
en-
vironments other
than
after
yes/no
questions)
Is
to
exhibit
on
the
part
of
its
producer
an
understanding
that
an
-
extended
unit
of
talk
is
underway
by
another,
and
that
it
is
not
yet.
or
may
not yet
be
(even
ought
not
yet
be),
complete
.
It
takes the stance
that
the
speaker
of
that
extended
unit
should
continue
talking,
and
in
that
continued
talking
should
con-
tinue
that
extended
unit
.
'tJh
huh',
etc
.
exhibit
this
under-
standing,
and
take
this
stance,
precisely
by
passing an
oppor-
tunity
to
produce
a
full
turn
at
talk
.
When
so
used, utter-
ances
such
as
'uh huh'
may
properly
be
termed
'continuers'
.
Note
that
the sorts of issues
mentioned
earlier
as
arising
with respect
to
the
'signalling
attention
and
understanding'
accounts
bear
differently
here.
(a)
For
talk
-
in
-interaction
whose
turn
-taking
organizatio
n
wakes
possible
-completion
-of
-one
-speaker's
-talk
a
place
wher
e
another
can
start
up
a
next
turn,
it
is
structurally relevant
at
such
places
for
parties
to
display
their
understanding
of
the
current
state
of
the
talk
.
For
example,
as
Sacks
pointed
out
years
ago, participants
sometimes
begin
a turn
by
produc-
ing
an
'uhm'
just
after
the
possible
completion
of a prior
turn,
then
pausing,
and
then
producing
a
turn,
rather
than
just
delaying the
start
of
their
turn
until
they
are
'ready'
.
They
may
be
understood
to
proceed
in this
fashion
precisely
in
order
first
to
show
their
understanding
of
the
current
state
of
the
talk
and
their
stance
toward
it
(i
.e
.
'a
prior
turn
is
over
.
It
in
an
appropriate
occasion
for
a
next
turn
.
I
will
produce
one'),
in
some
-
Independence
of the
actual
productlQn
of the turn
they
eventually
produce
.
So
also
in
it
relevant
for
parties
to
display
their
understanding
.
when
appropriate
.
that
an
extended
turn
is
underway,and
to
show
their
inten-
tion
to
pass
the
opportunity
to
take a
turn
at
talk
that
they
might
otherwise
initiate at
that
point
.
(b)
'Uh
huh's,
etc
.
as continuers
do
not
merely
claim
an
understanding
without displaying
anything
of the
understand-
ing
they
claim
.
The
production
of
talk in
a
possible
turn
position
which
is
nothing
other
than
'uh huh'
claims
not
only
'I
understand
the
state
of the
talk',
but
embodies
the
under-
standing
that
extended
talk
by
another
is
going
on
by
dectin-
log
to
produce
a
fuUer
turn
in that position
.
It
does
not
claim
understanding
in
general,
but
displays
a
particular
understanding
through
production
of
an
action
fitted to
that
understanding
.
1
1
(c)
Except
for
the
limited
set
of
behavioral
productions
that
are
used
to
do
'continuers',
it
is
not the case
that
any
in-
stance
of
an
indefinitely
extendable
set
of
utterances
would
achieve
this
outcome
or
do
this
job
.
Most
other forms of
talk
Would
be
full
turns
in
their
own
right,
rather than
ways
of
Passing
the
opportunity
to
produce
such
turn,
and
would
82
/
Emanuel
A.
Schegloff
fall
precisely
thereby
to
display
understanding
of,
or
respect
for,
an
extended
unit
still
in
progress
.
The
'continuer'
usage
is
most
readily
illustrated
by
data
in
which
cleat'
marking
of
the
end
of
the
extended
unit,
or
dis-
course,
is
provided,
and
until
the
occurrence
of
which
the
'in
-progress,
character of the
talk
is
clearly
visible
.
Among
the
ways
in
which
such
marking
may
be done
are
the
several
ways
of
announcing,
at
the
beginning
of the
unit,
the
sort
of
thing
that
will
be
its
possible
end
.
For
example,
there are
story prefaces
(cf
.
Sacks
1974)
which
may
characterize
the
sort
of
event
the
forthcoming
story
is
about
(for
example,
'a
funny
thing
happened
.
..
'),
such
that
the
unit
will
not be
possibly complete
until
such
an
event
has
been
mentioned
.
and
may
be
over
at
the
end
of
its
mention
.
Or
there
are
'prelimi-
naries
to preliminaries'
(Schegloff 1980)
in
which
an
'action-
type'
Is
projected
(like
'question'
in
'Can
I
ask
you
a
ques-
tion?')
as
that
to
which
preliminaries
are
leading
;the
prelimi-
naries
may
then
be
.developed
as
an
extended
discourse
(e
.g
.
a
description,
a
story, etc
.)
until
such
an
action
is
done
(e
.g
.
such
aquestion
in
asked)
as
these
preliminaries
could
be
leading
up
to
.
Several
instances
are given
in
(4)
and
(5)
.
(4)
1
:
I've
listen'
to
all
the things
that
chu've
said, 

an'
I
agree
with
2

you
so
much
.
3
:
NOW
.
-
4
'
I
wanna
ask
you
something,
S
:
I
'wrote
a
latter
.
8
(pause)
7
A :
Mhhm,
S
: 
T'the
governor
.
9
A : 
Mhhm
:
:,
10
:
-telling
'Lm
what
I
thought
about
i(hh)iu!
11
(A)
:
(Sh
:
:
:!
)
12
:
+
Will
I
get
an
answer
d'you
think,
13
A
:
Ye
:s,
(BC,
Red
;
190)
(5)
1 
B:
-
Now
listen
.
Mister
Crandall,
Let
me
ask
you

this
.
2
 A
cab. You're
standing
onna
corner
.
I 
heardlub
3
talking
to
a
cab
driver
.
4
A
:
Uh
:
:uh
S  B
: 
lJh
was
it
-
uh
was
a
cab
driver,
wasn'
I'?
6A:
Yup,
7 
B :
Now,
yer
standing
onna
corner
.
S 
A :
Mm
hm,
9  B
: 
I
live
up
here
in
Queens
.
10
 A
:
Mm
hm,
11
 B
:
Near
Queens
Boulevard
.
Discourse
as
an
Interactlonal
Achievement
1
83
12
A
:
Mm
hm,
13
B
:
I'm
standing
on
the corner
of
Queen
s
Boulevard
a:
:nd
14
uh
:
:m
( 
)
Street
.
15
A:
Right?
16
B
:
Uh
.
I
- a
cab
comes
along
.
An'
I
wave
my
 
arm,
"Okay,
17
1
wancha
I
wancha."
You
know,
18
A:
Miflin,
19
B:
IJh::m,
I'm
waving
my
arm
now
.
Here
in
 
my
living
room
.
20
hhhh
!
I
21
A
:
heh
heh!
22
B
;
A
;nd
uh,
he
just
goes
right
on
me
.
23
A
;
Mm
tun
.
24
B
;
A
;
;nd
uh
-two
;
:,
three:, (
.)
about
three
blocks
.
25
beyond
me,
where
-
In
the
direction
I'm
going,
there
26
Is
a
cab
stand
.
27
A:
Minhm,
28
B
;
Uh
-there
is
a
hospital,
(0
.')
uh
a
block
 
(0
.?)
up,
29
and
there
is
a
subway
station,
right
there
.
30
A
:
Mm
tuu
.
31
B
:
Uh
now
I
could
'ye
walked,
the
three
or
four
blocks,
32
to that
cab
stand,
33
A:
Mm
hm,
 
I
34
B :
 
Bud
I,
had
come
out-of
where
I
was,
 
right
there
35
on
the
corner
.
38
A
:
Right?
-
37
B
:
- 
Now
is
he not
suppose'
tuh
stop
fuh
me?
38
A
:
If
he
is
on duty
,
(W-C
.
Red
:
191-193)
Note
that
after
the
projection
of
a
question
upcoming,
the
re-
cipient
of the
extended
talk
connes
himself
almost
entirely
(the
alternatives
are
touched
on
below)
to
contlnuers
-
-'uh huh',
'mm
hmm',
'right',
and
the
like,
until
a
question
is
asked
(of
the
sort
analyzably projected; not
just
any
subsequent
ques-
tion
;
not,
therefore,
the
one
at
line
5
In
(5)
.
The
extended
unit
then
being
completed,
and
a
determinate
action
being
called for
by
the question,
the
recipient
of the
discourse
addresses
himself
to
the question.
The
same
form of
utter-
ance
may
be produced
(for
example,
the
'yes'
at
line
13
In
(4)),
but
in
this
sequential
environment
it
is
a
full
turn,
rather
than
passing
one
.
8
I
Emanuel
A.
Schegloff
What
will
constitute
the
end
of an
extended
spate
of
talk
is
not
always
named
or characterized
as
it is
in
the
aforemen-
tioned
forms
;
still
it
is
regularly
readily
recognized
by
the
participants
.
Sometimes,
however,
misunderstandings
occur,
and
a
continuer
produced
to
display
an
understanding
that
an
extended
unit
is
in
progress
and
is
not yet
completed
thereby
displays
a
misunderstanding,
as in
(6)
(taken from
the
same
corpus
of
telephone
calls
to
a
radio
talk
show,
as
was
the
source
of
(4)
and
(5)
.
(6)
1
B;
Tids
is
in reference to
a
call,
that
was
made

about
a
2 
month
ago
.
3
A
:
Yesair?
4
B:
A
woman
called,
uh
sayin
she
uh
signed
a
contract
for
5 
huh
son
who
is
-
whowas
a
minuh
.
6
A:
Mm
hm,
7
B:
And
she
claims
inna
contract
.
there
were

things
given,
8
and
then
taken
away,
in
small
writing
.
9 
((pause))
10
A
:
Mm
hm
11
B:
lJh,
now
meanwhile,
about
a
month
ehh
no
 
about
two
weeks
12
before
she
wade
the
call
I
read
in,
I
read
or

either
13
heard
-uh
I
either
read or
hoid
onna
television,
where
14
the
judge
.
hadda
case
like
this .
15
A
;
Mhhm,
16
B:
And
he
got
disgusted
an'
he says
'I'
-
he's
sick
of
these
17
cases
where
they
give
things
in
big
writing,
an'
take
18
'em,
an'
take
'em
away
in small
writing
.
19
A:
Mhhm,
20
B: Xi'
'e
claimed
the
contract
void
.
21
A:
Mhhm,
22
B
:
Uh
what
I
mean
is
it
c'd
help
this
woman
that
called
.
23
You
know
uh,
that's
the
reason
I
called
.
(BC,
Gray,
74
-75)
At
line
21
.
A
produces
another
in
the
series
of continuers
that
have
helped
propel
B's
telling
;
this
one,
it
turns
out,
is
'mistaken',
for
the
caller
had
apparently
intended
'An'
'e
claimed
the
contract
void'
to
be
the
end
-
-perhaps
hearable
as
'a
solution'
for
the
woman
to
whose
earlier
call
he
refers
.
12
It
is
worth
noting
that
'trouble'
around
the
end
boundary
of
discourse
units
need
not
be
understood
as
'cognitive'
or
Discourse
as
an
Interactional
Achievement
1
85
processing
error;
it
can
be
the
vehicle
for
thoroughly
de-
signed
Interactional
effects
(cf
.the
discussion
of
reengage-
went
of
turn
-by
-turn
talk
at
emergence
from a story
in
Jefferson
1978)
.
These
instances
allow
me
to
remark
on
several
additiona
l
points
which
way
provide
some
sense
of
the
interactional
tex
-
ture
involved
here
.
1.
Note
that
after
a
continuer,
the
speaker
of the
extende
d
unit
may
'do
the
continuing'
In
various
ways
(and
it
shoul
d
be
underscored
that
this
locus
of
talk
should
be
investigate
d
precisely
for
the
work
of
'doing
continuing')
.
In (4),
the
first
continuers
are
followed
by
increments
to
the turn-uni
t
(sentence)
already
in
progress;
In (5),
some
continuers are
followed
by
Increments
to
the
prior
sentence
(for
example
,
lines
1O
-l1)
others are
followed
by
starts
of
new
sentences
.
(for
example,
lines
12
-13)
;
still
others are
followed
by
wha
t
could be
counted
as
new
sentences
by
virtue
of
their
gram
-
matical
Independence,
or
as
Increments
to
the
prior
by
virtue
of
their
linkage
by
conjunction
-by
by
just
such
a
token
as
mark
s
'continuation'
(for
example,llnes
22-28)
.
In
this
respect
.
then,
there
in
no
major
differentiation
between
sentences
an
d
multi
-sentence
units
or
discourses
;
the
same
mechanism
ca
n
engender
an
elaborated
version of
the
former
or the
latter
.
2.
Note
that
the
bits
of behavior
produced
by
the
recipient
of
the
extended
talk
vary.
Two
points
may
be
advanced
here.
First
.
even
when
little
other
than
continuer
usage
is
involved,
the
tokens
employed
for
it
vary
.
I
have
referred
to
'uh
huh'
.
'yeah'
.
etc
.
throughout
this
paper,
and
have
not
addressed
myself
to
the
differences
between
these
tokens
.
I
note
here
only
that
the
availability
of a
range
of
tokens
may
matter
less
for
the
difference
of
meaning
or
usage
between
them
(if
any)
than
for
the
possiblUty
thereby
allowed
of
varying
the
compo-
sition
of
a
series
of
them
.
Use
in
tour
or
flve
consecutive
slots
of
the
same
token
may
then be
used
to
hint
incipient
disinterest,
while
varying
the tokens across
the
series,
whet-
ever
tokens
are employed,
may
mark
a
baseline
of
interest
.
Second,
In
some
of
the
positions
at
which
some
sort
of
con
-
tinuer
in
relevant
(as
may
be shown,
for
example,
by
th
e
speaker
withholding
further
talk
until
one
is
produced,
as
in
(4),
Unes
5
-7,
or
(8),
lines
8
-11),
the immediately
precedin
g
talk
may
be
such
as to
invite
some
sort
of
'reaction'
asid
e
from,
Instead
of,
or
in
addition
to
the
continuer.
And
on
e
does
flnd
throughout
extended
units
-
-especially
stories
-
markersmarkers
of
surprise
('Really?'),
assessments
('oh
my',
'wow',
'You're
kidding',
'isn't
that
weird',
'wonderful',
etc
.),
an
d
the
like
.
In
the
fragments
I
have
cited,
we
may
note
the
laugh in
(5)
at
line
21,
and
in
(4)
the
laugh/assessment
/
"
xpletjve
at
line
11
.
Note
in
the
case
of the
latter
that
it
foUow
a
selection
of
idiomatic
phrasing
that
indicates
'scold-
ing'
(and
this
has
already
been
reported
as
directed
to
a
high
Political
ofcial),
and
its
last
word
is
delivered with a
laugh
86
I
Emanuel
A
.
Schegloff
token
as
well
.
13
in the case of
(5),,
note
that
at
lines
16-17,
the
teller
'packages'
the
telling
in
a
very
dramatic
format
with
exaggerated
salt-quotation,
which
could
have
been
designed
to
engender
a
more
forthcoming
appreciation
than
this
'unp
tunm'
provides
."'
Note,
then,
that
although
B
does
continue
talking
after
the
continuer,
here
she
does
not
continue with the
ex-
tended
unit
that
was
'in
progress',
but
shifts
from
a
descrip-
tion
of
the
events
being
told
about
to
a
description
of the
cur-
rent
scene
of the
telling,
using
the
recurrence
of
'waving
the
arm'
as
the
bridge
.
The
description
appears
designed
to
underscore
'incongruity'
and
to
elicit
a
response
to
it
.
but
even
the
first
effort
at
this
fails
to
get
a
response
('I'm
wav-
ing
my
arm
now
.')
;
she
then
adds
another
(she
could
have
resumed
the story)
to
underscore
the incongruity
even
further
('Here
in
my
living
room').
to
which
she
appends
a
laugh token
as
weU
.
This
time
she
does
get
a
response
of the
sort
she
has
apparently
been
after
.
(Note
:
one
is
tempted
to
write
'of
a
sort
fitted
to
the
character
of
her
talk',
but,
of
course,
it
is
precisely
the
assessment
of
the character of
her
talk
which
is
at
issue
in
the
sort
of
response
A
wakes
or
withholds
.
It
may
be
suggested
that
the
mechanism
by
which
a
series of
same
continuer
tokens
displays
incipient disinterest
involves
the
availability
of tokens of
surprise,
special
interest
.
assess-
ment
etc
.
"
the
nonpro4uction
of
which
shows
the
recipient
not
to
be
finding
in
the
talk
anything
newsworthy,
Interesting,
or
assessable
.
Varying
the
continuer
tokens
may
mask
the
ab-
sence
of
othertypes
of
response
token;
using
the
same
one
continuously
may
underscore
it
.
)
The
general
point
I
want
to
make
here
is
that
the operation
of continuers
and
of the
other
bits
of
behavior
produced
by
recipients
in
the
course
of,
or
rather
in
the
enabling
of,
ex-
tended
talk
or
discourse
by
another,
is
designed
in
a
detailed
way
to
fit
to
the
ongoing
talk
by
the
teller,
and
'to
fit'
may
involve
either
'cooperating
with
what
that
talk
seems
designed
to
get,
or
withholding
;
both
of these
are
fitted to
the
details
of the
locally
preceding
talk,
and
cannot
be
properly
under-
stood or
appreciated
when
disengaged
from
it
.
When
disen-
gaged,
there
is
no
way
of
telling
that
the
'ann
limm'
at line
18
in
(5)
Is
not
only
a
continuer,
but
is
possibly
withholding
a
laugh
;
and
without
that,
one
may
not
be
in
a
position
to
under-
stand
why
the
teller
next
abandons
the story
for
a
description
of
her
telling
posture
.In
brief,
disengaging
the
listener
be-
havior
from
its
local
sequential
context not only
undercuts
the
possibility
of
understanding
what
it
to
doing
;
it
can
remove
an
important
basis
for
understanding
what
is
going
on
in
the
dis-
course
itself
.
The
preceding
discussion
having
ended
with
an
account of
some of
the
Interactional
texture
in
particular
data
fragments,
it is
in
point
to
recall
that
the
concern
of
this
section
is
to
see
what
more
general
assertions
can
defensibly
be
put
forth
Discourse
as
an
Interactional
Achievement
/
87
to
characterize
what
tokens
Uke
'uh huh'
may
be doing.
One
I
have
suggested
is
the
usage
as 'continuer'
.
The
continuer
usage
rests
on
the
observation
that
'ub
huh'
.
etc
.
passes
the
opportunity
to
do
any
sort
of
fuller
turn
at
all,
on
the
grounds
that
an extended
unit
is
already
in
pro-
gress
.Note.
however,
that,
were
a
fuller
turn
done
.
It
would
be some
particular
type
--it
would
be of
some
particular
form,
and
would
be
doing
some
particular
action
or
actions
.
In
passing
the
opportunity
to
do
a
fuller
turn,
a
participant
therefore
is
also
passing
the opportunity
to
do
something
in
particular
-
-the
opportunity
to
dowhatever
might
have
rele-
vantly
been
done
at
that
point.
We
just
discussed
a
case
in
which
an
'minhin'
was
alternative
to
a
laugh
;
but
we
clearly
cannot
say
that
'uh
huh',
etc. i
s
generally
a
way
of
with-
holding
laughter
.
because
there
is
no
way
of
showing
that
doing
laughter
is
generally
relevant,
and
if
something
cannot
be
shown
to
be
relevantly
present,
than
it
cannotbe
rele-
vantly
absent,
or
withheld
.
Of
course,
laughter
is
not
gener-
ally
relevant,
It
was
relevant
in
the
case
I
have
discussed
be-
cause
the
other
party
did
something
to
make
it
relevant,
and
that
is
why
oneneeds
the
local
sequential
environment
-
-tosee
what
the
other
parties
have
done
that
makes
some
sorts
of
next
actions
relevant,
which
'uh
huh'
may
be
displaying
the
withholding
of
.
The
question
is
.
are
there
any
kinds
of
actions
whichhavesome
kind
of
'general
relevance'
in
conver-
sation,
bywhich
in
meant
that
they
are
not
made
relevant
by
the
particulars
of
someone's
immediately
preceding
talk
or
be-
havior?
There
is at
least
one
candidate
.
One
kind of
talk that
appears
to
have
quite
ageneral
po-
tential
provenance
is
whet
has
elsewhere
(Schegloff
.
Jefferson,
and
Sacks
1977)
been termed
'other
-initiated repair'
or
'next-
turn
repair
initiation'
.
A
variety
of
.
constructional
formats are
used
to
do
the job
of
initiating
the
remedying
of
some
problem
of
hearing
or
understanding
the
just
prior
talk
of
another
--
several
of
the
WH
-question
terms,
such
as 'who', 'what'
.
etc
.,
as
well
as
'huh',
partial
(and
sometimes
full)
repeats
of
prior
turn,
partial
repeats plus
one
of the
question
words,
and
others (pp
.
367
-369).
It
appears
that
there are no
systematic
exclusion
rules
on
the
possible
relevance of
next
-turn
repair
Initiation
in
any
possible
turn
position
.
Although next
-turn
repair
initiation
is
generally withheld
until
after
completion
of
the
turn
in
which
the
trouble-source occurred
.
It
appears
cor-
rect
to
say
that
such
repair
initiation
is
regularly
potentially
relevant
after
completion
of
any
unit of
talk
by
another
.
15
Its
use
exploits
its
positioning
-
-next
after
the
unitin
which
the
trouble
-source
occurred
.
If
it
is
the
case
(Schegloff
.
Jefferson,
andSacks
1977:363)
that
any
talk
can
be
a
trouble-
source,
then
'after
any
talk'
can
be
a
place
for repair
to
be
initiated
on
it
.
Speakers
can
look
to
the
moments
after
some
unit of talk
to
find
whether
repair
on
that talk is
being
initi-
ated
;
indeed,
speakers
who
will
be
continuing
can
leave a
88
I
Emanuel
A
.
Schegloff
moment
of
nontalk
for
such
repair
to
be
Initiated
if
the
talk
just
produced
is
to
be
treated
by
others
as
a trouble-source.
Then
'uh huh',
nods,
and
the
like,
In
passing
the
opportunity
to
do
a
full
turn
at
talk,
can
be
seen
to
be
passing
an
op
V06
to
initiate
repair
on
the immediately
preceding
talk
.
Note
that,
if
tokens
such
as
'uh huh'
operate
to
pass
an
opportunity
to
initiate
repair,
the
basis
seems
clear
for
the
ordinary
inference
that
the
talk
into
which
they
are interpo-
lated
is
being
understood,
and
for
the
treatment
in
the
liter-
ature
that
they
signal
understanding
.
It is
not
that
there
in
a
direct
semantic
convention
in
which
'uh
huh' equals a
claim
or
signal
of
understanding
.
It
is
rather
that
devices
are
available
for
the
repair
of
problems
of
understanding
the
prior
talk,
and
the
passing
up
of
those
opportunities
.
which
'uh
huh'
can
do,
is
taken
as
betokening
the
absence
of
such
problems
.
Further,
the
use
of
other
-Initiated
repair
as
one
way
of
pre
-indicating
the
imminent
occurrence
of
disagreement
(Schegloff
.
Jefferson,
and
Sacks
1977
:380)
suggests
why
'ub
huh's
and
the
like
can
be
taken
as
indications
of
agreement
with the
speaker
of
an
ongoing
extended
unit
.
For
if
dis-
agreement
were
brewing, then
opportunities
to
initiate
repair
would
supply
a
ready
vehicle
for
the
display
and
potential
de-
flection
of
that
disagrement
.
Passing
the
opportunity
to
raise
problems
of
understanding
may
be
taken
as
indicating
the
ab-
sence
of
such
problems
.
It
may
also
be
taken
as
indicating
the
absence
of
that
which
such
problems
might
have
por-
tended
-
-disagreement
-
-
and
thus
be
taken
as
indications
of
agreement
.
It
must be
noted,
however,
that
there
is
a
difference
be-
tween
this
usage
and
the
continuer
usage
.
It
was
noted
earlier
that
with
regard
to
the
'current
state
of
the
talk'
.
'uh
huh'
does
more
than
claim
an
understanding,
but
embodies
it
in
particulars
and
acts
on
it
.
With respect
to
the
under-
standing
of,
and
agreement
with,
what
a
prior
speaker
has
said
and
done,
'uh
huh'
Is
merely
a
claim
of
understanding
.
Such
a
claim
may
turn
out
to
be
incorrect
;
and
passing
one
opportunity
to
initiate
repair
is
compatible
with
initiating
re-
pair
later
.
The
status
of 'ub huh'
as
an
indication
of
under-
standing
or
agreement
is
equivocal
in
a
way
in
which
its
status
as
a
continuer
is
not. as
participants
who
have
relied
on
it
will
have
discovered
and
regretted.
In
this
section,
1
have
tried
to
formulate
what
appear
to
me
to
be
the
only
two
general
characterizations
that
can
be
sus
.-
tamed
when
applied
to
singular,
particular,
situated
instances
of
vocalizations
such
as
'uh
huh'; a
usage
as
continuer
and
a
usage
to
pass
an
opportunity
to Initiate
repair
.
For
the
rest,
the
treatment of
them
in the
aggregate,
separated
from the
talk
immediately
preceding
them,
loses
what they
are
doing
.
Perhaps
more
germane
to
the
official
topic
of
this
Georgetown
University
Round
Table, along
with
that
is
lost
the
character
Discourse
as
an
interactional
Achievement
/
89
of the
ongoing
talk
during
which
they have been produced
.
Thereby
our
understanding
of
discourse
is
weakened
.
I
close
with several observations
on
this
theme
.
5
.
Among
the
themes
I
have
stressed
most
strongly
is
that,
at
least
in
conversation,
discourse
must be
treated
as
an
achievement
.
There
is
a
real,
recurrent
contingency
concern-
ing
'who
should
talk
now'
;
the
fact
that
someone
continues
is
an
outcome
coordinatedly
achieved
out of
that
contingency
.
There
is
a
real,
recurrent
contingency
concerning
what
who-
ever
-gets
-to
-talk should
talk
on
;
the
fact
that
the
same
speaker
who
talked before
talks
again
and
folks
more
of
the
same
thing
in
an outcome
achieved
out
of
this
contingency
.
(they could
have
gone
on
to
repair
what
preceded
;
they
could
have
paren-
thesized
into
a
comment
about
their
talking
;
they
could
have
'touched
-off'
Into
something
entirely
different,
etc
.)
.
Once
it
has
happened
that
'a
speaker
continues'
(for
exam-
ple
'a
teller
continues
his story'),
that
appears
entirely
'natural'
;
we
lose
sight
of
what
were
contingent
alternatives
;
they
do
not
become
'ex
-alternatives'
or
'alternatives
-not-
taken'
;
they
simply
jiaappear
.
and
leave
the achieved
outcome
hi
the splendid
isolation
of
seeming
inescapability
.
For
ana-
lysts, this
is
a
great
loss
.
Good
analysis
retains
a
sense
or
the
actual
as
an
achievement
from
among
possibilities
;
It
re-
tains
a
lively
sense
of the
contingency
of
real
things
.
It
is
worth
an
alert
.
therefore,
that
too
easy
a
notion
of
'discourse'
can
lose
us
that
.
.
If
certain
stable
forms
appear
to
amarge
or
recur
in
talk,
they
should
be
understood
as
an
orderliness
wrested
by
the
participants
from
interactional
contingency,
rather
than
as
automatic
products
of
standardized
plans.
Form,
one
might
say,
is
also
the
distillate
of
action
and/In
interaction,
not only
its
blueprint
.
If
that
in
so,
then
the
description
of
forms
of
behavior, forms of
discourse
(such
as stories)
included,
has
to
include
interaction
among
their
constitutive
domains,
and
not
just
as the stage
on
which
scripts
written
in
the
mind
are
played
out.
NOTES
My
appreciation
to
the
Netherlands
Institute
for
Advanced
Study
in
the
Social
Sciences
and
Humanities
(NIAS)
for
time
to
reflect
on
some
of the
matters
discussed
here,
while
I
was
a
Fellow
during
t978
-1919,
and
to
Anita
Pomerantz
and
Michael
Lynch
for
useful
discussion
.
1
.
The
behavioral
vehicles
for
interaction
between
'per-
former'
and
'audience'
may
vary
among
various
'slngle
speaker'
settings,
but
the
fact
of
interaction
is
certainly
not
limited'to
the
academic
lecture
.
Max
Atkinson
(private
communication)
ha
been
exploring
it
in
political
speeches
in
Great
Britain
.
90
I
Emanuel
A.
Schegloff
2.
See,
for
example,
the
paper
by
Marjorie
GoodwIn
(1980)
.
These
themes
are
relevant
not
only
for
discourse
units,
but
for
'sentences'
as
well
.
Cf. Charles
Goodwin
(1979)
.
3.
Of
course,
not
every
occurrence
of
one
of these
vocali-
zations
is
an Instance of the
usage
I
am
concerned
with
;not.
for
example,
occurrences
which
follow
so
-called
'yeslno
ques-
tions'
.
4
.
Once
again,
not
all
utterances
of
'first'
or
'first
of
all'
are
list
-initiating,
although they
do commonly
project
some
form
of
extended
al,
if
only
by
indicating
that
before
an
already
relevant
action,
something
else is to
be
done,
as in
the
follow-
ing
segment
;
Vie
:
I
knowwho
didit
.
James
;
You
know
who
didit
,
I
Vie
:
Yeeah
,
Vie
:
Ye:s
.
James
:
Who
wuzzit
.
t
)
Vic
:
+
First
of a:
:
:11,
un
Michael
came
by:,
.
.
.
(US,
33)
S
.
The
alternative
is
adding
to
the turn
unit
already
pro-
duced,
which
can
then
be
recompleted,
as in
the
following
:
Anne
;
Apparently
Marcia
went
shopping
for
all
these
things
.
(L
W
Anne
;
Becuz
uh
;
(0 .5)
Lash
didn't
seem
t'kno
:w,
which
kid/Id
-
-
(Poet
-Party,
1,
5)
6
.
On
the
possibilities
discussed
in this
paragraph
.
c
f
.
Sacks
et
*1
.
(1914:704,
709,
715)
.
7.
Cf
. note
16
.
8.
Kendon
does
describe
another
use
of
such
interpolations
-
-&
'point
granting'
use
.
9.
In
Fries'
materials
from
telephone
conversation,
and
in
Dtttman
and
Llewellyn's
experimental
format
(1967;348).
the
parties
are not
visually
mutually
accessible,
and
this
remark
is
not
in
point
.
10
.
As
it
happens,
a
number
of
Yngve's
instances are of
this sort
;
cf
.
Yngve
1970
:574
.
11
. Cf.
Fragment
6,
lines
20-23,
and
the
discussion
in
note
12
.
12
. Note
that
B's
'what
I
mean.
.
.'
shows
an
orientation
to
'having
been
misunderstood'
.
He
does
not
go
on
to
say
he
means
to
help the
woman
and
this
was
the
reason
for his
call
;
he
uses
a
repair format
to indicate that
this
is
what
he
meant
before,
which
was
not
understood
by
A.
as
displayed
by
the
'mh
hm'
which
indicates
waiting
for
more
to
come
.
This
bears
Discourse
as
an
Interactional
Achievement
/
91
on
the
remark
earlier
in
the
text that
continuer
tokens
dis-
play
an
understanding
of the
current
state
of the
talk,
and
do
not
merely
claim
an
understanding
.
It
is
the
displaying
of
what
understanding
their
producer
has
whichmakes
it
possi-
ble for
recipient
of
the
continuer
to find that
understanding
flawed
.
13
.
On
the
ways
in
which
a
laugh
token
can
solicit
a
re-
sponse
from
a
coparticipanc,
cI
.
Jefferson
(1979)
.
14
.
It
so,
then
the
'mm
hm'
may
be used
in
lieu
of,
or
to
display
the
withholding
of,
such
a
more
forthcoming
response,
a
possibility
further
examined
later
.
Note
too
that
'uh huh',
etc
.
can
be
delivered
in
an
indefinitely
extendable
range
of
ways
;
some
'uh
huh's
can
mark
surprise, appreciation,
assess-
ment,
etc
.
15
.
Indeed,
it
can
be
relevant
after
a
suspected
talk unit
by
another,
as
exchanges
such
as the
following
show
:
(Silence
)
A
:
Huh
?
B
:
I
didn't
say
anything
.
(EAS
:
FN)
16
.
In
this
respect,
'uh
huh',
'mm
hmm'
"
nods,
and
the
Uko
are
specifically
alternatives
to
utterances
such
as
'huh?',
'what?',
'who?',
and
the
like,
rather than
being
comembers
of
a
category
such
as
'back
-channel
communications',
as in
Yngve
(1970)
andDuncan
and
Flake
(1971)
.
On
the
other
hand,
'uh
huh',
etc
., in
being
alternatives
to
repair
initiation,
are
in
a
sense part
of
the
organizational
domain
of
repair
.
In
writing
in
the
text
of 'passing the
opportunity
to
do
a
full
turn
at
talk'
.
I
appear
to
be
joining
the
consensus
re
-
ported
on,
and
joined
by,
Duncan
and
Flake
(1977:203)
that
'back
-channel
actions,
in
themselves,
do
not
constitute
speak-
ing
turns'
.
However
.
I
do
not
believe
that
(a)
this
question
should
be
settled
on
conceptual or
definitional
grounds
;
(b)
the
various
components
included
in
the
term
'back
-channel'
tare
identically
on
this
question
;
or
(a) positions
on
the
turn-
Status
of 'uh huh'
are
Invariant
to
the occasion
for
the issue
being
posed
.
I
can
here only
suggest
the basis
for
this
stance
.
Consider
the
fragments
in
(I)
through
(iii
.
(1)
D:
But
listen
tuh
how
long
-
R;
I
I
In
other
words,
you
gotta string
up
the-
you
gotta siring
up
the
colors,
is
that
it?
(KC
-4
.
37)
(II)
 R
:
Hey
;
;, the
place looks
different
.
F
;
Yea
;
:hh
.
9
I
Emanuel
A
.
Schegioff
K
:
Ye
have
to
see
all
ou
r
new
-
1
3
It
does?
R
:
Oh
yeah
.
(KC
-4,2
;
cf
.
Sacks
at
al .,
p.
720)
(111)
1 B:
hhh
And
he's
going
to
make
his
own
paint
-
ings,
2 A:
mm
hmm
.
3
B:
And
-
or
I
mean
his
own
frames
.
4
A
:
yeah
.
(SBL
:
1, 1,
12-11)
Note
first
that
in
both
(I)
and
(U), talk
which
requests
clarifi-
cation (In
(II))
or
repeats
and
solicits
confirmation
(in
(1)),
which
are
two
types
of
back-channel
for
Duncan
and
Fiske,
win
out
in
floor
fights,
though,
according
to
Duncan
and
Fiske,
it
is
a
consequence
of
beck
-channels
not
being
turns
that
in-
stances
like
these are not
even
counted
by
them
an
simul-
taneous
turns.
In
my
view,
the
issue
of the turn-status of
some
utterance
should
be
approached
empirically
.
I
.e
.
do
the
parties
treat
it
as a turn;
In
(i)
and
(ii)
.
clarification
talk
is
so
treated
.
I
believe
much
talk
of
this
sort
is
treated
by
participants
as
having
full
turn
status
.
However,
other
sorts
of
vocalization,
such
as 'uh
huh',
are
not so
treated,
as
Dun-
can
end
Fiske note,
at
least
with
respect
to
simultaneous
talk
and
its
resolution
.
When
the issue
is
a
different
one,
however,
a
different
posi-
tion
may
be
warranted
.
In
(iii),
for
example,
'paintings'
in
line
1
is
an
error,
which
is
corrected
at
line
3
by
its
speaker
.
This
correction
is
undertaken
after
the
recipient
has had
an
opportunity
to
do
so,
and
has
passed
.
With respect
to
the
organization
of
repair
and
its
interactional
Import,
it
can
matter
that
B's
self-correction
follows
a
passed
opportunity
for
A
to
initiate
repair
.
A
silence
by
A
in that position
may
well
have
called
attention
to
the
presence
of
a
repairable
;
the
'mm
hm',
in
specifically
not
doing
so,
is
doing
something
.
'Mm
hm'
is
more
than
'not
a
turn'
;
with
respect
to
the
repair
issue
.
It
is
very
much
like
one
.
Accordingly
.
It
seems
appropriate
to
me
that
the
turn-status
of
'uh
huh'
etc
. b
e
assessed
on
a
case
-by
-case
basis,
by
reference
to
the
local
sequential
environment,
and
by
refer-
ence
to
the
sequential
and
interactional
issues
which
animate
that
environment
.
REFERENCES
Dtttman,
Allen
T
.,
and
Lynn
G
.
Llewellyn
.
1067
.
The
pho
-
nemic
clause
as
a
unit
of
speech
decoding. Journal of
Per
-
sonality
and
Social
Psychology
6
.341
"
-349
.
Discourse
as
an
Interactional
Achievement
/
93
Dlttman,
Allen
T
.,
andLynn
G
.
Llewellyn
.
1968
.
Relation-
ship
between
vocalizations
and
head
nods
as
listener
re-
sponses
.
Journal
of
Personality
and
Social
Psychology
9.79-84.
Duncan
.
Starkey
and
Donald
W
.
FIske
.
1977
. Face-to
-face
Interaction
:
Research
.
methods,
and
theory
.
Hilladale,
N
.J
.:
Lawrence
Erlbaum
Associates
.
Fries,
Charles
C
.
1952
.
The
structure of English.
New
York
:
Harcourt,
Brace
.
Goodwin,
Charles
.
1979
.
The
interactive
construction
of a
sentence
in
natural
conversation
.
In
.
Everyday
language
:
Studies
in
ethnomethodology
.
Edited
by
George
Psathas
.
New
York
:
Irvington
Publishers
.
Goodwin,
Marjorie
.
1980
.
Processes
of
mutual
monitoring
im-
plicated in
the
production
of
description
sequences
.
Socio-
logical
inquiry
50
.303
-317
.
Jefferson.
Gail
.
1978
.
Sequential
aspects of
storytelling
i
n
conversation
.
In
:
Studies
In
the
organization
of
conver
-sational
interaction
.
Edited
by
Jim
Schenkein
.
New
York
:
Academic
Press.
Jefferson,
Gail
.
1979
.
A
technique
for inviting
laughter
and
its
subsequent
acceptance
/declInatIon
.
In
:
Everyday
lan-
guage
:
Studies
in
ethnomethodology
.
Edited
by
George
Psattuis
.
New
-York
:
Irvingion
Publishers
.
Kendon
.
Adam
.
1967
.
Some
functions
of gaze
direction
in
social
interaction
.
Acta
Psychologica
26
.22-63.
Sacks,
Harvey
.
1974
.
An
analysis of
the
course
of a
joke's
telling
in
conversation
.
In
.
Explorations
in
the
ethnogra-
phy
of speaking
.
Edited
by
R
.
Baumann
and
I
.
Sherzer
.
Cambridge
:
Cambridge
University
Press.
Sacks
.
Harvey,
and
Emanuel
A
.
Schegloff
.
1979
.
Two
prefer-
ences
in
the
organization of reference
tO
persons
in
con-
versation
and
their
Interaction
.
In
:
Everyday
language
:
Studies
in
ethnome(hodology
.
Edited
by
George
Psathas
.
New
York
:
£rvington
Publishers
.
Sacks,
Harvey, Emanuel
A.
Schegloff,
and
Gail
Jefferson
.
1974
.
A
simplest
systematica
for
the
organization
of
turn
-taking
for
conversation
.
Lg
.
50
.896-735
.
Scheglof(
.
Emanuel
A
.
1968
.
Sequencing
in
conversational
openings
.
American
Anthropologist
70
.1075
-1095
.
Schegloff,
Emanuel
A.
1973
.
Recycled
turn beginnings
.
Public
lecture
at
Linguistic
Institute,
University of
Michigan,
Ann
Arbor
.
Mimeo
.
Schegloff,
Emanuel
A.
1980
.
PrelIminaries
to
preliminaries
:
'Can
I
ask
you
a
question'
.
Sociological
inquIry
50
.104
-152.
Schegtof(,
Emanuel
A
.,
Gail
Jefferson,
and
Harvey
Sacks
.
1977
.
The
preference
for
self
-correction
in
the
organiza-
tion
of
repair
in
conversation
.Lg.
53
.361-382
.
Yngve,
Victor
.
1970
.
On
getting
a
word
in
edgewise
.
In
:
Papers
from
the
Sixth
Regional
Meeting,
Chicago
Linguisti
c
Society
.
Chicago
:
Chicago
Linguistic
Society
.
567
-577
.
}
... This system crucially depends on the notion of the Transition Relevance Place (TRP), which is an opportunity in the current speaker's utterance at which a listener can, but is not obligated to, take over the role of speaker. Even short feedback-like turns, such as 'hmm', known as backchannels (Yngve, 1970) or continuers (Schegloff, 1982), are precisely timed to occur at TRPs. Importantly, TRPs are not a function of a speaker's intentions but a consequence of the turn-taking mechanism itself. ...
... Beyond the basic mechanisms, cultural variations also play a role in how turn-taking unfolds. While the fundamental mechanisms of turn-taking, such as the cues for taking or passing on turns (Stivers et al., 2009), are largely universal, cultural norms can shape the timing and style of these transitions (Schegloff, 1982). Therefore, understanding both the culture-invariant and culture-specific components of turn-taking is crucial for developing dialogue systems that are not only responsive but also adaptable across diverse cultural contexts. ...
Preprint
Full-text available
Turn-taking is a fundamental mechanism in human communication that ensures smooth and coherent verbal interactions. Recent advances in Large Language Models (LLMs) have motivated their use in improving the turn-taking capabilities of Spoken Dialogue Systems (SDS), such as their ability to respond at appropriate times. However, existing models often struggle to predict opportunities for speaking -- called Transition Relevance Places (TRPs) -- in natural, unscripted conversations, focusing only on turn-final TRPs and not within-turn TRPs. To address these limitations, we introduce a novel dataset of participant-labeled within-turn TRPs and use it to evaluate the performance of state-of-the-art LLMs in predicting opportunities for speaking. Our experiments reveal the current limitations of LLMs in modeling unscripted spoken interactions, highlighting areas for improvement and paving the way for more naturalistic dialogue systems.
... In spoken communication, the role of feedback is crucial to streamline conversations and convey the interlocutor's state-ofmind, e.g., to signal attention, understanding, or attitude [1,2,3]. Feedback can be expressed in different forms through voice, face and body [4,5,6]. ...
... The transcript was missing in some cases; these samples were taken away when text embeddings were considered. Here we used a 80/10/10% split for training, validation and test, respectively 1 . ...
... Seen in this light, "transition spaces are (…) organizationally strategic" (Schegloff 1996: 97). The strategic use of linguistic resources at points of potential turn transition may occur in the form of a "rush-through" (Schegloff 1982) or "abrupt-join" (Walker and Local 2004), or in the form of the phenomena described below. This concept of the "possible TCU" as a resource rather than a linguistic format is also supported by Selting's (2000) understanding of TCUs without TRPs, that is, TCUs that are designed for their non-final position in multi-unit turns. ...
... When used in responses to informings and tellings, yeah/no may be deployed as acknowledgement tokens accomplishing the work of continuers (Jefferson 1984(Jefferson , 2002Mazeland 1990;Schegloff 1982) that treat the co-participant's turn as still ongoing and align with the telling activity in progress. Moreover, in these contexts, yeah has also been found to display incipient speakership, projecting a transition from recipiency to the adoption of a speaker role. ...
Chapter
Full-text available
This collection of original papers illustrates recent trends and new perspectives for future research in Interactional Linguistics (IL). Since the research program was started around the turn of the century, it has prospered internationally. Recently, however, new developments have opened up new perspectives for interactional linguistic research. IL continues to study the details of talk in social interaction, with a focus on linguistic resources and structures of verbal and vocal interaction in bodily-visible interactional settings. Increasingly, though, it embraces methods supported by new technology and broadens its data and research questions to applications in teaching, therapy, etc. The volume comprises three parts with 14 contributions: (1) Studying linguistic resources in social interaction; (2) Studying linguistic resources in embodied social interaction; and (3) Studying social interaction in institutional contexts and involving speakers with specific proficiencies.
... La percepción e interpretación de los hechos dentro de la narración se complementan de la orientación espacio temporal y locativo, de la descripción y de la evaluación del actuar y comportamiento de los participantes, de la necesidad de establecer planos para que se entiendan las secuencias narrativas (Bamberg y Marchman, 1991;Schegloff, 1982;Segal y Scott, 1991). El proceso cognitivo de la narración obliga a construir una jerarquía de esquemas mentales, y para lograr la cohesión de la producción narrativa, se depende de las formas lingüísticas que van enlazando las ideas dentro del discurso. ...
Thesis
Full-text available
El incremento del lenguaje posterior a la etapa preescolar es lento, sutil, y a veces difícil de cuantificar (Nippold, 2006); no obstante, se pueden obtener evidencias de los cambios tardíos a través del análisis de muestras del lenguaje de niños entre los 6 y los 12 años de edad, incluso, de jóvenes mayores (15 años). Con la edad los niños muestran un cambio en su complejidad gramatical debido a la experiencia con el lenguaje y a que cambian sus necesidades comunicativas (Coloma, Peñaloza, Reyes Fernández, 2007). La presente investigación se dio a la tarea de analizar ese cambio en la complejidad a través de producciones narrativas de niños que se encontraban en etapas tardías de desarrollo. El foco de análisis son los conectores, como elementos cohesivos que permiten la coordinación y subordinación de cláusulas, en producciones generadas por niños de 1º grado (6-7 años) y 6º grado (11-12 años) de Primaria y 3º grado de Secundaria (14-15 años). Los resultados sugieren que el desarrollo del lenguaje establece el escenario de la competencia del mismo. Además muestran que los conectores son utilizados como indicadores funcionales para organizar el texto y crear una determinada perspectiva textual, para enfatizar puntos decisivos, rebatir puntos anteriores, etc. Los hallazgos de este trabajo contribuyen al conocimiento sobre el desarrollo del lenguaje en etapas tardías pues provee información relevante sobre la progresión y adecuación de elementos discursivos: conectores interclausales en muestras narrativas. (Palabras clave: coordinación y subordinación de cláusulas, etapas tardías de desarrollo, producciones narrativas, conectores interclausales)
... By co-construction, I specifically refer to the dialogical processes through which interlocutors jointly construct new knowledge during their argumentative interactions. The issue of co-construction through dialogical interactions has already been widely addressed within the fields of social and developmental psychology (Cole, 1985;Doise & Mugny, 1984;Rogoff, 1990), conversation analysis (Goodwin, 1979(Goodwin, , 1995Heritage, 1984;Schegloff, 1982), and linguistic Complimentary Copy anthropology (Duranti, 1985;Ochs, 1988;Schieffelin & Ochs, 1986). As stated by Jacoby and Ochs (1995, p. 171): "Co-construction does not necessarily entail affiliative or supportive interactions. ...
Chapter
Full-text available
This chapter focuses on the family and school context to examine how parents and children and teachers and students contribute interactionally to managing their argumentative discussions. Within a data corpus comprising 30 video-recorded meals of 10 Swiss and Italian families (sub-corpus A) and 16 video-recorded lessons of two courses-one at the undergraduate level and one at the graduate level-in Developmental Psychology (sub-corpus B), argumentative discussions were selected for qualitative analysis by adopting the pragma-dialectical model of a critical discussion and the Argumentum Model of Topics (AMT). Despite the differences in roles, age, and competencies between parents and children and between teachers and students, the findings of this study indicate that in the family and the school context, managing their argumentative discussions is a co-constructed dialogical process. In particular, by engaging in argumentative discussions, parents and teachers accept the commitment to clarifying to children and students the reasons on which their standpoints are based. Children and students, in turn, encourage parents and teachers to advance arguments to justify their standpoints by asking questions. In conclusion, this study shows how argumentative discussions in the family and school context open a shared space for parents and children and teachers and students to reason together. Future research might consider the observations and the subtle qualitative analyses of argumentative discussions in the family and school
... 78), sino como una especie de evaluación (p. 85) (Schegloff, 1982) que dirige la atención a lo previamente dicho por la PE para tratarlo como un «hallazgo» que corresponde ser notado, soslayando, en este caso, su carácter de «teoría» (línea 21) y de «intención» (línea 25), dándole así un estatus noticioso. ...
Article
Con el propósito de extender el conocimiento sobre cómo opera la desinformación en el periodismo, se estudia un caso de desinformación que atraviesa televisión y prensa. Aplicando la teoría y métodos del Análisis de la Conversación, esta investigación analiza la entrevista política de televisión en vivo (en adelante, EPTV) que produjo la desinformación que fue luego ampliamente difundida por diversas notas periodísticas. Desde una perspectiva interaccional, la investigación da cuenta de lo ocurrido en la EPTV, identificando prácticas conversacionales que llevan a la producción de desinformación, con particular énfasis en las prácticas periodísticas. Por su parte, las notas periodísticas muestran, de forma indirecta, las repercusiones de la desinformación en la audiencia, a través de cómo diversos medios, con distintas afinidades ideológicas (izquierda, derecha), recibieron, comprendieron, y, diseminaron la desinformación producida en la EPTV. Asimismo, se identifican tanto las implicancias teóricas como las prácticas. Entre ellas, se destaca que la persona que hace la entrevista (PHE) pregunta a la persona entrevistada (PE) por experiencias vicarias de las cuales ésta no tiene cómo dar cuenta, así como que la PHE se convierte en coautora de las respuestas a sus propias preguntas. También se constata que la PHE recurre a prácticas conversacionales que son características del chisme y que resultan ajenas al habla institucionalizada que caracteriza (normativamente) a la EPTV.
... The recipients' actions show us that they orient towards Strong Finals as an indication that there is more to come. Ann does this by inserting a sequence with a misplacement marker, and the other friend, Dorit, does this by providing the response token hm (lines 5 and 8), serving as a 'continuer' (Schegloff 1982, not to be confused with Christensen's 'continuer intonation', 2010), and thereby supporting Melissa's storytelling project. ...
Article
Full-text available
This article presents structural and interactional aspects of Strong Finals, a prosodic feature characterised by lengthening, increased volume, and non-falling intonation on word-final syllables. Interactionally, Strong Finals support five types of action: listing, projecting a description, stating conditions, asking questions, and announcing reported speech. In general, Strong Finals project that there is more to come, and this ‘more’ may in some cases be provided by either participant. Strong Finals are often found in multi-speaker settings, where they assist speakers in taking the floor or changing the topic. The article’s descriptions are based on recordings of natural spoken interaction in linguistically diverse areas in Aarhus, Denmark. Here, a new urban dialect has developed like other urban dialects that have been described in Copenhagen and other North Germanic cities. Strong Finals are a local phenomenon, however, and are not found in the Copenhagen studies.
... Interactants typically mark horizontal and vertical transitions implicitly, via words like uh-huh, yeah, right, all right, or okay. These markers have been previously studied as backchannels [4], continuers [5], acknowledgment tokens [6], or discourse markers [7]. [3] proposed to regroup these words into the category of project markers, suggesting that different lexi-* These authors share last authorship. ...
Conference Paper
Full-text available
Joint activities (e.g. building a LEGO model) unfold in a hierarchy of subprojects. Navigating them implies horizontally elaborating on a subproject (placing one block) and vertically moving to a new subproject (next block). Interactants coordinate horizontal and vertical transitions with project markers (okay, yeah). We suggest that vertical vs. horizontal transitions are distinguished both lexically and acoustically. We predicted that acoustic features of identical markers used for different transitions (okay-vertical vs. okay-horizontal) would exhibit more dissimilarity than markers used for same transitions (okay-vertical vs. okay-vertical). We used MFCC-based dynamic time warping to measure dissimilarity between vocalisations and analysed them with a Bayesian regression model. We find that Vietnamese speakers use both lexical and acoustic cues to mark transitions, and paired same-horizontal markers are acoustically more similar than same-vertical and different-transition markers.