Glossary of transcript symbols with
an introduction
Gail Jefferson
Although I’d probably rather transcribe than any do any other part of the work
(analyzing, theorizing, lecturing, teaching, etc.), the one thing I’d rather not do
is talk about transcribing. It’s not a topic. You might as well talk about typewriting.
Transcribing is just something one does to prepare materials for analysis,
theorizing, etc. Do the best you can, but what is there to talk about?
On the other hand, there might be something to talk about if we compare
a 1964 transcript by Harvey Sacks,
(1) [Sacks GTS trans:1964]
A. I started work at a buck thirty an hour and he said if I work a month you
geta buck thirty five an hour and every month there be a raiseT.
Howd you get the job?
A. I just went down there and asked him for it.
T. Last week you were mentioning something about the fact that youA.
I got lost in one job, yeah.
T. Got lost in it, and your fatherC.
He know your father? Yeah.
[ A. Sure he knows my father, but my father’s got nothing to do with it.
[ C. Ok he gave you the job.
A. No, he’s got nothing to do with it. Huh uh, My father’s not buying beer
anymore
with a transcript that I did in 1985.
 Gail Jefferson
(2) [GTS:I:2:3:R:1-5:3-4]
Ken: I started workin etta buck thirty en hour
(0.4)
Ken: en’e sid that if I work fer a month: yih getta buck,h ·h thi[rty ↓fi:ve=
(Dan): [((sniff))
Ken: =’n hour en (·) ev’ry month he uh ( ) he rai[ses you ]◦
( )◦
]
Dan: [How’dju]g e t th]e jo:b,
(1.0)
Ken: ↑I js wen’ down there’n ↓a:st eem for it
(1.8)
Dan: ◦
Cz◦
la:st week you were mentioning something about th’ fa:ct ◦
thet you
↓u[h◦
Ken: [I got ul u. -got (·) lost in one jub=↓Yea:h.
(0.5)
(i t bo:th[ered) ]
Dan: Got lo:st innit ↓e[r (y’r fa:th[e r ) ]◦
( )◦
Ken: [ w h h h h [h h h h]h h
Al: [Dz ’ee ] know yer father?
(0.2)
Al: ↓Yah.
(0.6)
Ken: Sure ’ee knows my father [ b’t my f a t h e r’s g’t] nothina ] do with it.
Al: [( ) they gave you] th’ jo:b,]
(0.7)
Ken: No: he’s got nothin d’do w’th it. Huh-uh, my fa(h)ather’s not buyin beer
innymo:re
Fragment (1) is concise, and readable; fragment (2) is a nightmare.
Or, for example, we could look at a recently produced, succinct and readable
transcript from Stanley I. Kutler’s Abuse of Power (1997:253),1
(3) [AbPow:253]
((re the “seamy things” Hunt says he did for the White House Plumbers))
DEAN: I don’t know the full extent of it,
PRESIDENT NIXON: I don’t know about anything else.
→ DEAN: I don’t know either, and I [laughs] almost hate to learn some of these
things.
and ask why I feel compelled to produce something like this:2
Glossary of transcript symbols 
(4) [Jeff:Canc:40:10-20]
Dean: I ↑don’t kno:w th˙e (·) full extent ↓’v it.↓
(0.7)
Dean: ◦
↓Uh:::eh◦
(0.9)
Nixon: ◦
I don’noo◦
’bout anything else exchh
e[pt
Dean: → [I don’t either in I: ◦
w’d (h)als(h)o
→ hhate tuh learn [some a’] these thi]ngs. ·hh·hh·hh·hh
Nixon: [W e l l ] y a : h ]
(0.2)
Dean: So ↑That’s,hhhh that’s that situation.
Why put all that stuff in? Well, as they say, because it’s there. Of course there’s
a whole lot of stuff “there,” i.e., in the tapes, and it doesn’t all show up in my
transcripts; so it’s because it’s there, plus I think it’s interesting. Things like
overlap, laughter, and ‘pronunciational particulars’, (what others call ‘comic
book’ and/or stereotyped renderings), for example. My transcripts pay a lot of
attention to those sorts of features.
What good are they? I suppose that could be argued in principle, but it
seems to me that one cannot know what one will find until one finds it, so
what I’ll do is show some places where attention to such features turned out to
be fruitful.
First of all, we can compare bits of fragments (1) and (2), and then do the
same with fragments (3) and (4).
Starting off with fragments (1) and (2), I’ll focus on the most irritating
segment of the latter, and see what might be gotten from it. Recalling three
lines from fragment (1):
A: I got lost in one job, yeah.
T: Got lost in it, and your fatherC.
He know your father? Yeah.
and the jumble from fragment (2):
Ken: I got u.l-got (·) lost in one jub=↓Yea:h.
(0.5)
(i t bo:th[ered) ]
Dan: Got lo:st innit ↓e[r (y’r fa:th[e r ) ]◦
( )◦
Ken: [ w h h h h [h h h h]h h
Al: [Dz ’ee ] know yer father?
(0.2)
Al: ↓Yah.
 Gail Jefferson
What is so clearly to be made of this bit in fragment (1) is that the therapist
mentions patient A’s father, and the mention of A’s father triggers a question by
patient C:
T: ...and your fatherC.
He know your father?
In fragment (2) the issue is drastically obscured, first of all by the sheer difficulty
of reading through all that stuff, now including some sort of whoosh
of breath by Ken (A), but mainly by the addition of an alternative hearing for
“your father”, i.e., “it bothered”. Surely that could be resolved by looking at the
context: We have the therapist, (Dan/T), mentioning the word “father”, which
prompts one patient (Al/C) to ask the other (Ken/A) a question about that
father, starting up before the therapist finishes whatever he was going to say.
The alternative is rather improbable, i.e., that Dan has said something
about Ken’s having been “bothered”, and it just so happens that at that moment
Al decides to cut in with a tangential question about Ken’s father. Logic
plumps for “father”; so much so that I only kept the alternative in as a matter
of principle. I wished that I could unhear it. But I do hear it; it would be irresponsible
to leave it out even though it not only messes up the transcript but
it’s so improbable that I’m embarrassed by it.
But on thinking about it, yes, it’s more than likely that the “father” alternative
is what’s happening with Al, but that doesn’t tell us what Dan said. In our
materials we sometimes can clearly hear one speaker saying something, and
find that a coparticipant somehow has heard something else. Once in a while,
one or the other mentions the mishearing. For example:
(5) [Frankel:GS:X]
((re a plant Alan is trying to sell to Nell))
Alan: → Still growing. It’s got buds ’n everything else on it.
Nell: Oh has i:t?
Alan: → Buds. No[t bugs. ]
Nell: → [Oh bu:d]s. I thought you said bugs.
Alan: No. I don’t see any bugs. It might have, but I can’t see any
It’s a phenomenon. And it very likely happens more often than we have access
to in our materials, because people don’t always, or even routinely, mention
the fact that they did a mishearing (sometimes they don’t realize it, sometimes
they do realize it but it doesn’t seem to matter, etc.).
What the “bothered” alternative does is to raise the possibility that such
a thing has happened here: Dan is saying something about Ken having been
Glossary of transcript symbols 
“bothered”, which Al hears as his saying something about Ken’s “father”, which
inspires the question he then and there asks (‘then and there’ in fine detail,
i.e., starting up after the first syllable of what he hears as the trigger-word,
“fa:th[er”).3
I’ll just mention one other difference between the two fragments. One of
them shows an ‘and’ while the other shows an ‘or’ (pronounced “er”).
A. I got lost in one job, yeah.
T. → Got lost in it, and your father–
versus
Ken: I got ul u.-got (·) lost in one jub=↓Yea:h.
(0.5)
(it bo:thered)
Dan: → Got lo:st innit ↓e[r ( y’r fa:ther ) ◦
( )◦
Ken: [w h h h h h h h h h
So what, “and” or “or”. Well, “and” is accepting a proffered version of what
happened, and augmenting it: ‘You got lost in it and it bothered...’ / ‘You got
lost in it and your father...”. If it’s “or”, then while repeating the proffered version
of what happened may acknowledge it as reasonable, the “or” projects
an alternative version; not that you “got lost in it”, but that “...it both[ered
( )...” / but that “...your fath[er ( )...”. So, the difference between “and” and
“or” is a difference between two polar activities: One transcript shows the therapist
accepting, the other shows him rejecting, the patient’s version of what
had happened.
Turning to fragments (3) and (4), focusing on the occurrence of laughter.
In fragment (3), we are shown some talk in progress, in the course of which
Dean “[laughs]”:
PRESIDENT NIXON: I don’t know about anything else.
DEAN: I don’t know either, and I [laughs] almost hate to learn some of these
things.
In fragment (4) we are shown some details of those two utterances:
Nixon: ◦
I don’noo◦
’bout anything else exchh
e[pt
Dean: [I don’t either in I: ◦
w’d
(h)als(h)o hhate tuh learn some a’ these things.
For one, fragment (4) suggests that Nixon is going on to mention something
else he knows about, i.e., we now have the word ‘except’. And we have Dean
starting up within that word, at “exce...”, where after, Nixon stops. One thing
that might be happening here is that Dean hears, in “exce...”, the word ‘except’
 Gail Jefferson
forming up, and starts to talk at that point. This ‘recognitional- response’ is a
not-uncommon phenomenon. Here are a couple of cases.4
(6) [Rah:II:11:R]
Jessie: ◦
two pihleece cah:rs’d◦
stopped outsi:de.=
Ada: =eeYe::[s?
Jessie: → [·h An’ that whether he thow:t thet I ed’n
ac[cidn’t] [ohr someth]ing...
Ada: → [I : : : :]:[ k n e o : w ]
In fragment (6) above, just as Jessie starts to say the dreaded word “ac[cident”,
Ada starts her comforting “I:::: kno:w”. And in fragment (7) below, just as Mr.
Bryant is pointing out that with the credit note his firm is offering Miss Sokol,
their services won’t “co[st” her anything, she starts up with her acceptance of
his offer:
(7) [SCC:DCD:37:R]
Bryant: → ...but ·hh if you’ve gohtta credit neote ih weon’t
co[st you anything anyw]ay.
Sokol: → [ Wul owright the:n, ]
Sokol: That’s faiuh.
Similarly, with “exce...”, Dean may hear the word ‘except’ forming up. Hearing
that, he may hear that Nixon is starting to mention something else, knowledge
of which Dean doesn’t want to be burdened with. It may be that Dean
moves then and there to stop any possible revelations by cutting in on the
alerting word ‘except’, prior to its completion, with an agreement that specifically
ignores its projection of further things (i.e., his “I don’t either” targets
Nixon’s initial proposal, “I don’t know about anything else”), and then goes on
to announce his unwillingness to know any more.
Then there is the laughter. It appears that there is not merely laughter in
the utterance with which Dean may be declining to hear what Nixon started
to say, but that Nixon, in projecting such talk, himself produces something
that might be taken for laughter. That is, fragment (4) shows that he produces
the word “except” with some breathiness and an unvocalized mid-vowel:
“exchh
ept”. Again, here is a possible case of a recurrent phenomenon: A recipient
of someone’s utterance may treat something in that utterance as laughter,
and respond with laughter of his own. Here’s a simple case:5
Glossary of transcript symbols 
(8) [Schenkein:II]
Bill: Dju watch by any chance Miss International Showcase las’ night?
Ellen: n:No I didn’[I w’z reading myBill:
→ [You missed a really great pro(h)[gram.
Ellen: → [O(hh)h i(h)t wa(hh)s?=
Ellen: =ehh heh heh heh!
Here, just after a plosive breath occurs in the course of Bill’s utterance, Ellen
produces a laughter-loaded response.
In fragment (4), Dean may be doing something similar, but less directly –
and less transparently – reciprocal. That is, he does not immediately start to
laugh, but first produces a bit of talk. He’s not, then, to be heard as slavishly
laughing just because the President is laughing. As well, the work he’s doing
with that bit of talk may be enhanced by his not only ignoring the word
“except”, but declining to reciprocate the laughter with which the problematic
word is being produced.
Furthermore, his delayed laughter, while perhaps taking Nixon’s prior
laughter as a warrant (i.e., it is all right for him to laugh since the President has
just laughed), can, by its delay, be heard for the work it may independently be
doing, e.g., as tempering his declination to hear what the President was perhaps
about to tell him.6
Yet another point can be made. It is possible that the transcriber of fragment
(3), by not attending the details of the laughter, but just noting that it
occurred, misheard the talk in which that laughter occurs, i.e., fragment (3)
shows Dean saying “I [laughs] almost...”, while fragment (4) shows him saying
“I: w’d (h)als(h)o...” (in standard orthography, “I would also”).
In an early paper on laughter (Jefferson 1985), I proposed that laughter
can make an utterance less (or un-) intelligible, and may possibly be used for
that feature. In the case at hand, Dean’s laughter, while not being deployed for
that purpose, may have made the utterance that I hear as “I would also...” less
intelligible, resulting in its being heard by another transcriber as “I almost...”.
Here’s another sort of thing that can be noticed in a comparison of fragments
(3) and (4). In the former, Dean, shown of course in standard orthography,
is saying “...I [laughs] almost hate to learn some of these things.”
In the latter, among other pronunciational particulars, he’s shown as saying
“tuh learn”.
In an early paper on this issue (Jefferson 1983), I point out that “the sort
of ‘comic book’ orthography I use (e.g., for ‘What are you doing?’, ‘Wutche
doin’?’) is considered objectionable in that it makes the speakers look ‘stupid’;
 Gail Jefferson
it seems to caricature them rather than illuminate features of their talk”, and
that “experts on phonetics such as William Labov, propose that someone who,
for example, says ‘dat’ instead of ‘that’, is not producing defective English but
is speaking correctly in his dialect, and thus should not be transcript-displayed
as producing an object which is commonly treated as defective.”
In that paper, as part of my defense of ‘pronunciational particulars’ I show
several fragments in which speakers of one or another ‘dialect’ can be found
to be varying their pronunciations; for example, a member of the California
motorcycle gang, the Hell’s Angels, produces “them” followed by “dere”
(‘their’):
(9) [KPFK:GJ]
Joe: I tell them right t’dere face ...
For example, a Bronx janitor produces several versions of ‘there’ (and ‘they’re’):
“theyuh”, “they’re”,“deyuh”, and “dere”, while consistently using “th” for (two
different versions of) ‘the’:
(10) [Frankel:USI:117:R]
Vic: We get in they:uh (0.5) en they’re uh (0.2) the tu- (·) u. (·) t-two ↓guys uh
deyuh, ’n me ’n James ↑Wal↓kuh’s dere ’n th’ broa:d is in th’ bed.
For example, A Philadelphia-Italian meatcutter uses both “dis” and “this”:
(11) [Goodwin:M:3-4]
Frank: → Fu(h)ck you. Dis g[uy is–
Joe: [Hey wait.
.
.
Frank: → I definitely ain’t goi(h)in’ ou(h)t with this chick again.
And, for example, in a fragment of a transcript I made of a tape collected by
Labov, we find both Mez, a member of a black teenage street gang, and B.J.,
a black social worker/ ex-street gang member, producing “d” and “th”; Mez
saying “Hey lookih dat”, “Who’s that. Who’s that punk right there.”, (and also
producing a vowel-begun version of ‘that’s’, “Ahz Davey More.”), B.J., saying
things like “y’all see these pictichiz...”, “y’know who dat is”, and “That’s Davey
More”.
Glossary of transcript symbols 
(12) [Labov:Jets:3-5]
Mez:→ Hey lookih dat.
.
B.J.: → Jo man, we gonna talk about fights man y’all see these pictchiz
→ up here man y’know who dat is,
Mez:→ Who’s that. Who that punk right there.
Mayall: Mothuhfuck[im!!
B.J.: → [A punk, [That’s–
Mayall: [He stinks!
Alex: [Shuddup Mayall!!
Leonard: [Man- eh heh! [khhkhkhkhkhkhkh[khkhkh
B.J.: [Davey More. [
Mez:→ [Ahz Davey More
B.J.: → Th-that’s Davey More en that’s- Benny Kid Paret.
I then go on to locate some possible systematicities and interactional phenomena
that can be found by looking at pronunciational details, concluding that by
omitting such particulars from our transcripts, we are obliterating a potentially
fruitful data base.
Now, that paper specifically focused on talk by people who stereotypically
mispronounce (or, as Labov has it, correctly-for-their-dialect pronounce) various
words, who can be found to do correct (or incorrect-for-their-dialect)
pronunciations, as well. But if we look at the talk of John Dean, surely an
epitome of WASP, middle class, etc. etc., we also find variation.
So, for example, in a fragment from Abuse of Power, Dean uses the word
“to”, three times within a short spate of talk:
(13) [AbPow:247-248]
⇓ ⇓
DEAN: It started with an instruction to me from Bob Haldeman to see if
we couldn’t set up a perfectly legitimate campaign intelligence operation over
at the Re-Election Committee.
PRESIDENT NIXON: Hm-hm. ⇓
DEAN: Not being in this business, I turned to somebody who had been
in this business, Jack Caulfield, who is, I don’t know if you remember Jack or
not. He was...
My transcript shows that on each of the three occasions of its occurrence,
Dean pronounces the word ‘to’ differently: “to”, “tuh”, and “dih”:
 Gail Jefferson
(14) [Jeff:Canc:7:27-8:9]
Dean: → ↑Started with (1.0) en instruction to me:: (0.9) from Bob ↓Haldeman.
→ (0.4) tuh see if we couldn’t set up ay perfectly le↓git’mate (0.3) campaign
intelligence operation over et the ReElection Committee.
(·)
Nixon: Mmhm,
(0.8)
Dean: → Not being in this business?=I turned dih somebuddy who: ·h had been
in this business: (·) Jack Cau’field who: w’z I don’ know if you r’member
Ja:ck er not he w’z...
We can at least note that the ‘correct’ pronunciation, “to”, coincides with a stress
on that word, while the two ‘incorrect’ pronunciations, “tuh” and “dih” occur
at points where the word is not being stressed. Also, we can at least account
for the occurrence of the ‘d’-begun “dih”, as conditioned by the preceding,
‘d’-ended word, “turned”. Who knows what other orderlinesses will emerge
as attention is given to such details?
Harvey Sacks often spoke of the possibility of “order at all points”, and in
one of his lectures (Sacks 1966 [1992:I:238]) tells us that “we’re dealing with
something real and powerful. And not just grossly powerful, like, it provides for
the rate of industrial development, but it provides for little tiny things that God
might have overlooked, perhaps.” Most of the things Sacks dealt with, while
not as gross as the rate of industrial development, were on a larger scale than
tracking the variations in a Bronx janitor’s pronunciations of ‘there’, or John
Dean’s pronunciations of ‘to’. But now and then, something in one of our more
detailed transcripts would catch his attention, and we’d be treated to “a little
something I find enormously amusing”, having to do with someone’s spelling
out their name in a traffic court, producing it as “M-a-u-e-r, (pause) h-a-n.”,
the point being that “pause distributions are kind of important...in this sense at
least: You can do them wrong.”, which led to a consideration of the “normative”
character of pausings (Sacks 1968 [1992:I:784). Or, for example, we’d be told
of “a fascination” he had with “[t]hings like ‘didje’ and ‘wanche”’, and his having
“noticed several differences between them”, which led to an exploration of
the phenomenon of “transformation” in actual talk (Sacks 1969 [1992:II:137–
139]). (I have a feeling that Sacks’ specifically formulating this sort of material
as something he found “enormously amusing”, or had “a fascination” with, was
dealing with their on-the-face-of-it utter trivialness, even for the sort of mundane
stuff he was known to work with. These “little tiny things” were perhaps
beginning to be just a bit too tiny for comfort.)
Glossary of transcript symbols 
But they’re “there” in the talk recorded on the tapes, and many of them
are captured in the transcripts that use the system explicated in the ensuing
glossary of transcript symbols. Some of them have led to the discovery of ranges
of orderlinesses; most of them are yet to be explored.
Notes
. In his introduction, Kutler says “I have edited the conversations with an eye toward eliminating
what I believe insignificant, trivial, or repetitious ... and often have omitted dutiful
choruses of agreement by those present unless I believed them particularly important. The
dialogue of innumerable uses of ‘right’, ‘yeah’, ‘okay’ often has been dropped ... The ‘uhs’
and ‘ahs’ usually have been eliminated”. (p. vii)
. This and subsequent fragments titled and referred to as “Jeff:Canc” are from a transcript
I did in October, 2000 of the March 21, 1973 ‘Cancer on the Presidency’ meeting between
Nixon and Dean.
. Another possibility it raises is that Dan did say “bothered”, that Al did hear it as such, and
that the sound- similarity of the first syllables of “both[ered” and “fath[er” triggered Al’s
question to Ken about “father”.
. For a discussion, and further cases, of this phenomenon, see Jefferson (1984: esp. 28–29).
In its way, fragment (2) is also a case of ‘recognitional’ onset of talk, if not specifically of
recognitional ‘response’.
. A version of fragment (8), and a discussion of this phenomenon can be found in Jefferson
(1979:82–83).
. Transcribing the Watergate materials, I’ve been struck by the delicacy of some of Dean’s
interactional work. I didn’t notice anything of that nature in this particular fragment until I
focused on it for this exercise, having chosen the fragment simply because it gave more detail
to the Abuse of Power version, “...and I [laughs] almost hate to learn some of these things”.
References
Jefferson, G. (1979). “A technique for inviting laughter and its subsequent acceptancedeclination”.
In G. Psathas (Ed.), Everyday Language: Studies in Ethnomethodology
(pp. 79–96). New York: Irvington.
Jefferson, G. (1983). “Issues in the transcription of naturally-occurring talk: Caricature
versus capturing pronunciational particulars”. In Tilburg Papers in Language and
Literature, 34, 1–12. Tilburg: Tilburg University.
Jefferson, G. (1984). “Notes on some orderlinesses of overlap onset”. In V. D’Urso & P.
Leonardi (Eds.), Discourse Analysis and Natural Rhetorics (pp. 11–38). Padua: Cleup
Editore.
 Gail Jefferson
Jefferson, G. (1985). “An exercise in the transcription and analysis of laughter”. In T. van
Dijk (Ed.), Handbook of Discourse Analysis, Vol. 3: Discourse and Dialogue (pp. 24–34).
London: Academic.
Kutler, Stanley I. (1997). Abuse of Power. New York: Touchstone.
Sacks, Harvey (1992). Lectures on Conversation, Volumes I and II, G. Jefferson (Ed.).
Cambridge, MA: Blackwell.
Glossary of transcript symbols
// Double obliques indicate the point at which a current speaker’s talk is
overlapped by the talk of another. ((No longer in use.1
))
Louise: ’N how t//all are you, Al,
Roger: How tall’r you Al.
[ A left bracket indicates the point of overlap onset. ((The currently-used
alternative to the double obliques. Note also, a change in descriptive
language.))
Louise: ’N how tall [are you, Al,
Roger: [How tall ’r you Al.
] A right bracket indicates the point at which two overlapping utterances
end, if they end simultaneously, or the point at which one of them
ends in the course of the other. It also is used to parse out segments of
overlapping utterances.
Louise: ’N how t[a l l u h r ] you ]↓A:]l,
Roger: [How tall ’r] ↓you] Al, ]
= Equal signs indicate no break or gap.
A pair of equal signs, one at the end of one line and one at the beginning
of a next, indicate no break between the two lines.
Maggie: . . .en ’e weighs about a hunnerd ’n thirdy five pounds.=
Ronald: =AAUUGH! WHADDA L-LIE!
The pair is also used as a transcript convenience when a single speaker’s
talk is broken up in the transcript, but is actually through-produced by
its speaker.
Pammy: Yeah well okeedoe=
Myra: =[Yeah.
Pammy: [I j’s thought I’d ask
1. The asterisk (*) was used by some transcribers to indicate termination of simultaneous
speech in conjunction with double obliques (//). (Ed.)
Glossary of transcript symbols 
In this case, Pammy’s utterance is produced as “Yeah well okeedoe I j’s
thought I’d ask”, with Myra’s “Yeah” starting up immediately upon completion
of “okeedoe”, and simultaneously with “I j’s. . .”.
A single equal sign indicates no break in an ongoing piece of talk, where
one might otherwise expect it, e.g., after a completed sentence.
Ehrlichman: . . . so I said I jis’ find that hard to ima↓gine.=Now
(0.4) ·p ↑since ↓then I’ve retained coun↓sel.
(0.0) Numbers in parentheses indicate elapsed time by tenths of seconds.
Al: . . .j’s be a lot’v (shh) lotta work- lotta hassle.
(0.2)
Al: =[Well,
Roger: [Well if yer goin’ t’ all that trouble,
– – Double dashes indicate a short, untimed interval without talk, e.g., a ‘beat’.
((no longer in use))
Vic: I’m intuh my thing, intuh my – – attitude
against othuh pih- ·hh
(·) A dot in parentheses indicates a brief interval (± a tenth of a second) within
or between utterances.
Mrs A: ’Ello:?
Guy: ’Ello is Curly there?
→ (·)
Mrs A: → Oo jis (·) ˙e-Who:?
Guy: Johnny?h An[sin?]
Mrs A: [Oo j]ist ↑a minnih,
_____
| Numbers in parentheses bracketing several lines of transcript indicate time
(0.0) elapsed between the end of the utterance or sound in the first bracketed
__|___ line and the start of the utterance or sound in the last bracketed line.
Mrs A: _____ Oo jist ↑a minnih,
(0.6)
Kid: (1.2) ( [ )
Mrs A: __|___ [It’s fer you dea:r,
____ Underscoring indicates some form of stress, via pitch and/or amplitude.
A short underscore indicates lighter stress than does a long underscore.
Ehrlichman: Well Dean has: uh:,h totally coop’rated
with the U.S. Attorney.
:: Colons indicate prolongation of the immediately prior sound. The longer
the colon row, the longer the prolongation.
 Gail Jefferson
Mike: ↑iYeh it’s all in the ↑chair=
Mike: =↓all th[at ◦
junk’s in the chair.◦
↓]
Vic: [W o : : : : : : : : : : ] ↓ : : : w : .=
Vic: =I din’ know that?
:__ Combinations of underscore and colons indicate intonation contours.
Basically, the underscore ‘punches up’ the sound it occurs beneath.
wo:rd If a letter preceding a colon is underscored, the sound represented
by that letter is ‘punched up’, i.e., an underscored letter followed by
a colon indicates an ‘up-to-down’ contour.
Kalmbach: Hi:.=
Ehrlichman: =How’r you:.
wo:rd If the colon is underscored, then the sound at the point of the
colon is ‘punched up’, i.e., a letter followed by an underscored colon
indicates a ‘down-to-up’ contour.
Emma: Is SA:M there with [yuh?]
Lottie: → [Y e : ]a h,=
Emma: → =Uh ha[:h,
Lottie: [Uh huh
wo:rd If underscoring occurs prior to the vowel preceding the colon, then
the entire word is ‘punched up’, i.e., the colon indicates prolongation
only; there is no mid-word shift in pitch.
Vic: ’M not saying he works ha:rd.
In multi-syllabic words, if the consonent is underscored, then all syllables
thereafter are ‘punched up’.
Ehrlichman: He said ˙e-I came dih you:,hh fr’m Mitchell,hh
en I sai:d,h uh↓ : Mitchell needs money?
Here, the first mention of “Mitchell”, with only the initial consonant
underscored, is produced with the entire word ‘punched up’, while
in the second mention, “Mitchell” with the underscored vowel, pitch
drops at the second syllable. Likewise, the entire word “money” with
only the initial ‘m’ underscored, is ‘punched up’.
↑↓ Arrows indicate shifts into especially high or low pitch.
Dan: ◦
Thet’s a good ↑ques↓tion.◦
(0.6)
Louise: ↑↑Thank ↓you.
Glossary of transcript symbols 
.,?? Punctuation markers are used to indicate ‘the usual’ intonation. (The italicized
question-mark [?] substitutes for the question-mark/comma of my
non-computer transcripts, and indicates a weaker rise than that indicated
by a standard question-mark.2
) These symbols usually occur at appropriate
syntactical points, but occasionally there are such displays as the
following.
Maggie: Oh I’d say he’s about what.=five three enna ha:lf?=aren’t
chu Ronald,
Sometimes, at a point where a punctuation marker would be appropriate,
there isn’t one. The absence of an ‘utterance-final’ punctuation marker
indicates some sort of ‘indeterminate’ contour.
WORD Upper case indicates especially loud sounds relative to the surrounding
talk.
Kalmbach: I returned it ’n went over the:re (·) tih↑da:y, (0.5) A::ND
uh (0.8) he said the ↑rea:son thet. . .
◦
word◦
Degree signs bracketing an utterance or utterance-part indicates that the
sounds are softer than the surrounding talk.
Leslie: But we were ↑very ↓sorry to hea:r (·) that uh (·)
◦
your mother◦
had (·) died is that ri:ght Phi[↑lip?
Philip: [Yeah.
...
Ava: B’t the point is Jessie don’t fehget no:w. ·h (0.3) eh:m
(·) ’e w’ so: close t’◦
Mart’n◦
wa:sn’t ↓ ’e.
* Asterisk
In some transcripts, the asterisk indicates percussive non-speech sounds,
e.g., as in the following fragment, a fist thumping a table.
Vic: BU(h)D I’M NO(h)T I(h)NTUH THA(h)*T! * *
In non-computer transcripts, the asterisk indicates ‘creaky voice’. (In computer
transcripts, I’ve stopped tracking ‘creaky voice’ and am using the
asterisk for another phenomenon.)
Emma: En ar air co.nditioner went out. comin’ ba:ck
so Go*:d.=
Lottie: =◦
O[h:: G*o:d.◦
Emma: [itEmma:
Oh:: God ih w’z hot. . .
2. The inverted question-mark (¿) is also used as a substitute for the question-mark/comma
by some transcribers. (Ed.)
 Gail Jefferson
t*, d* In early computer transcripts, an asterisk following a consonant replaces
the single sub- or superimposed dot which serves as a ‘hardener’ in my
non-computer transcripts.
Kalmbach: I w’jist (·) understa:nd thet* uh: you en I are dehabs’ooly
dihgether on tha:t,
Ehrlichman: No question about it*?=uh hHerb
In this case, while Kalmbach produces “jist” and “tha:t,” with the
American-standard, soft ‘t’, the ‘t’ in his “thet*” is crisp, dentalized, i.e.,
‘hard’.
Similarly, while Ehrlichman produces “about” with the soft ‘t’, the ‘t’ in his
“it*” is ‘hard’.
t, d In more recent computer transcripts, a boldface consonant replaces the
single sub- or superimposed dot which serves as a ‘hardener’ in my noncomputer
transcripts. The above fragment would now be shown as:
Kalmbach: I w’jist (·) understa:nd thet uh: you en I are dehabs’ooly
dihgether on tha:t,
Ehrlichman: No question about it?=uh hHerb
a.,˙e,ï When a single dot is not available, two dots over a vowel replace the single
sub- or superimposed dot which, as well as a ‘hardener’, serves as a
‘shortener’ in my non-computer transcripts.
Ehrlichman: ˙e-he:: told me::? . . .an:d uh,h ï-he sid ...
Here, while conceivably the ‘e-’ in “e-he” and the ‘i-’ in “i-he” could be
read as long sounds, “ee” and “eye”, the single dot over the ‘e-’ and the double
dots over the ‘i-’, confirm that those sounds are short. I don’t show the
sounds as “eh” and “ih” because they are more fleeting than those spellings
indicate.
The dots do an additional job in transcripts where I use non-standard orthography.
Many words get a range of oddball spellings, in keeping with
the range of pronunciations they are subject to. On occasion such a word
appears in its standard spelling. If that word carries the dot(s), it means
that while such a spelling could be the result of a lapse of transcriber concentration,
in this case the standard spelling does indicate the way the
word was pronounced.
Emma: En ar air co.nditioner went out. comin’ ba:ck
Here, while ‘air conditioner’ is routinely pronounced as ‘air c’nditioner’, it
is being given a fully formed vowel, shown as “co.nditioner”.
...
(a) Kalmbach: Ehm: I:’m uh scheduled fo. r ↑two duhmorrow afternoo:n.
.
.
Glossary of transcript symbols 
(b) Kalmbach: . . .he said the ↑rea:son thet wz: u.-fer the ca:ll w’z. . .
In this case, while at point (b) Kalmbach is shown pronouncing the word
‘for’ as “fer”, the dot below the ‘o’ in “for” at point (a) indicates that
it’s not that the transcriber had simply written the word in its standard
orthography, but that it is there fully pronounced as “for”.
(b) A parenthesized italicized letter replaces the parenthesized letter with a
sub- or superscribed degree sign which, in my non-computer transcripts,
indicates an ‘incipient sound’.
Emma: you couldn’ ev’n putcher hand ou:ts:I:de the CAR
ih jiz’(b)bu:rn.
...
Ehrlichman: But they- (·) th˙e(p) the point is. . .
whord An italicized ‘h’ appearing in such a word as ‘which’ ‘where’, ‘what’ ‘when’,
‘whether’, etc., indicates that while such words are often produced with the
‘h’ silent (as if they were the words ‘witch’, ‘wear’, ‘wen’, ‘weather’, etc.), in
this case the ‘h’ was sounded.
Ehrlichman: En I said well Joh:n what ’n the world er yih talking
↓about*.
.
.
Ehrlichman: See ↑what they’ve said duh Dean is...
While at one point in a conversation Ehrlichman pronounces the word
‘what’ with the ‘h’ sounded, at a later point, the ‘what’ is produced with
no ‘h’.
word An italicized letter replaces the sub- or superscribed degree sign which, in
my non-computer transcripts, indicates unvoiced production.
Ehrlichman:→ He said we:ll?=hmhh ˙e-I came dih you:,hh fr’m
Mitchell,hh en I sai:d*,h uh↓ : Mitchell needs money?
(Kalmbach):→ (◦◦
Right◦◦
)
Ehrlichman: Uh::: could*=uh we::: ca::ll Herb Kalmbach en ask
’im duh raise ↓some.
Kalmbach: → ◦◦
Yeah.◦◦
<word A pre-positioned left carat is a ‘left push’, indicating a hurried start; in effect,
an utterance trying to have started a bit sooner than it actually did.
This can be heard, for example, as a compressed onset of the utterance or
utterance-part in question. A common locus of this phenomenon is ‘self
repair’.
 Gail Jefferson
Ruth: Monday nights we play, (0.3) <I mean we go to ceramics,
...
Polly: Y’see it’s diff’rent f’me:. <eh f’(·) the othuh boy:s,
word< A post-positioned left carat indicates that while a word is fully completed,
it seems to stop suddenly.
Meier: Uh well I fel’ like my lef’ side of my (·) chest I c’d (·) mah
had a k- cramp<
– A dash indicates a cut-off.
Vic: He said- yihknow, I get- I get sick behind it.
> < Right/left carats bracketing an utterance or utterance-part indicate that the
bracketed material is speeded up, compared to the surrounding talk.
< > Left/right carats bracketing an utterance or utterance-part indicate that the
bracketed material is slowed down, compared to the surrounding talk.
·hhh A dot-prefixed row of ‘h’s indicates an inbreath. Without the dot, the ‘h’s
indicate an outbreath.
wohhrd A row of ‘h’s within a word indicates breathiness. In some transcripts the
‘h’s are italicized, in some not.
Colson: . . .a ghhuy wh(h)o’s olso totally loyal<
And in computer transcripts, the ‘h’s are sometimes also superscribed.
Colson: Th:e ↑thing thet worries me is the(p)(·) is the
phh
ossibility’v. . .
(h) Parenthesized ‘h’ indicates plosiveness. This can be associated with laughter,
crying, breathlessness, etc.
Jim: Don’t sound so (h)amp(h)itious fer Ch(h)rise’sake
(h)ih suh ·hh sou’ l(h)i’ yuh k(h)uh g(h)o tuh sleep
’n the pho(h)one.
...
Maggie: I j’st ran up th’ stai(hh)rs that’s wh(h)y I’m
huffing en puffing.
£ The pound-sterling sign indicates a certain quality of voice which conveys
‘suppressed laughter’ (various transcripts have other symbols, e.g.,
the Dutch guilder-sign which no longer exists).
Ken: ahh ha £I don’ know who’s payin’ fer thi(h)s I
↑think ih my fa:-ther.£
wghord A ‘gh’ stuck into a word indicates gutteralness. In some transcripts the ‘gh’
is italicized, in others, not.
Mike: Ah don’ think ’ee lives onna ↓groun’ flo↓ :h.
(0.3)
James: The: ghghroun’ flo’
Glossary of transcript symbols 
In this case, a speaker with phleghm in his throat is saying “the ground
floor”, with the word ‘ground’ heavily gutteralized.
( ) Empty parentheses indicate that the transcriber was unable to get what was
said. The length of the parenthesized space reflects the length of the ungotten
talk.
Mike: No.
(0.4)
Mike: ( ),
In the speaker-designation column, the empty parentheses indicate transcriber’s
inability to identify a speaker.
Roger: Paz’m z’m Miller Highlit*e.h
(0.5)
( ): hnh Yhehh
(word) Parenthesized words and speaker designations are especially dubious.
(Mike) [(Lee me alone.)]
Carol: [Mike I know yu]h’ love’ m
(blerf) Nonsense syllables are sometimes provided, to give at least an indication of
various features of the un-gotten material.
Nixon: Jerry sh’d talk to ↑Witnaw. (0.5) And uh: (·) jis brace
eem ’n tell ’im tih (·)(offih sebbatikiss). . .
(Ø) A nul sign indicates that there may not be talk occurring; that what is being
heard as possibly talk might also be ambient noise.
Nixon: _____ ◦
(Well ah’ll protect chu but* uh)◦
|
( Ø ): (0.7) (Okay.)
|
Nixon: __|__ ◦
(Thet uh)◦
thet’s that’s why:. (0.9) ◦◦
I◦◦
(0.5) can’t
let chu ↓go (0.2) go dow:n.
(( )) Doubled parentheses contain transcriber’s descriptions.
Ray : ehh-heh-heh-heh-heh-he:h-eh=
Maggie: =((dainty snort))
...
Vic: ((dumb slob voice)) Well we usetuh do dis