BRAIN AND LANGUAGE 55, 199–212 (1996)
ARTICLE NO. 0101

Disconnected Phonology: A Linguistic Analysis
of Phonemic Jargon Aphasia
ROBERT E. HANLON
Department of Neurology, Washington University School of Medicine
AND

JEROLD A. EDMONDSON
Program in Linguistics, University of Texas, Arlington, Texas
This paper reports on indications of the nature of the neurolinguistic connection
between phonological and lexical components of language, based on a case of phonemic jargon aphasia. Following bihemispheric embolic infarcts, the subject presented with severe fluent aphasia, characterized by fluent strings of phonemes, with
virtually no intelligible utterances. Despite nearly total jargonized output, the fundamental phonological processes of speech were largely intact. Specifically she demonstrated: (1) English phonotactics and English stress-timed rhythmic principles,
(2) aspirated stops word-initially and glottalized stops word-finally, (3) utterancefinal declination of pitch, and (4) stressed syllable vowel lengthening. Additionally,
regional-specific (Southern American English) phonological processes, including
monophthongization, in-gliding, and front vowel backing, were also preserved.
Overall, the investigation reveals an example of an intact phonological rule system operating on a grossly disturbed input (lexical representation).  1996 Academic
Press, Inc.

INTRODUCTION

Jargon utterances are commonly produced by fluent aphasics and subtypes
of aphasic jargon have been delineated. Alajouanine (1956) differentiated
three types of aphasic jargon, based on comparative preservation of linguistic
organization. Paraphasic jargon was characterized by fluent and grammatically intact output contaminated by semantic paraphasic errors. Asemantic
jargon consisted of generally intelligible speech with preserved syntax conThe authors thank Sara Green, MS, Neuropsychology Department, and Milton Thomas,
Department of Physical Medicine, Baylor Institute for Rehabilitation, for their assistance with
this study. Address correspondence and reprint requests to Robert Hanlon, Department of
Neurology, Washington University School of Medicine, Box 8111, 660 South Euclid, St.
Louis, MO 63110.
199
0093-934X/96 $18.00
Copyright  1996 by Academic Press, Inc.
All rights of reproduction in any form reserved.

200

HANLON AND EDMONDSON

taminated by neologisms. Finally, undifferentiated jargon consisted of unintelligible and totally meaningless phonemic sequences, with a tendency for
stereotypy and perseveration. Perecman and Brown (1985) adapted Alajouanine’s classification according to the presumed level of linguistic processing
from which a particular type of jargon is derived. Accordingly, Alajouanine’s
paraphasic jargon, characterized by semantic paraphasias was referred to as
semantic jargon. The presence of neologisms denoted neologistic jargon.
And phonemic jargon referred to meaningless strings of utterances, presumably originating at the phonological level.
Numerous reports of patients with neologistic and semantic jargon are
available in the literature, with regard to anatomical findings (Buckingham &
Kertesz, 1974; Kertesz, 1981; Kertesz & Benson, 1970; Lhermitte, Lecours,
Ducarne, & Escourolla, 1973), neurolinguistic analysis (Lecours & Rouillon,
1976), psycholinguistic analysis (Buckingham, 1987; Buckingham, Avakian-Whitaker, & Whitaker, 1978; Butterworth, 1979; O’Connell, 1981; Semenza, Cipolotti, & Denes, 1992), anosognosia (Lebrun, 1987; Weinstein &
Lyerly, 1976), and language therapy (Kotten, 1982). However, relatively few
empirical analyses of phonemic jargon are available, apparently due to the
comparatively infrequent nature of such jargon in Wernicke’s aphasia (Perecman & Brown, 1981; Cappa, Miozzo, & Frugoni, 1994).
In the case of K.S., a patient with phonemic jargon aphasia, secondary to
bilateral temporoparietal infarcts, Perecman and Brown (1981) described the
relative frequency distribution of phonemes. They found a markedly greater
frequency of /m/ and /b/ than normal and relatively higher proportions of
the phonemes /r/ and /s/. Anderson’s (1981) examination of the syntactic
boundary cues of K.S. revealed normal final vowel lengthening during spontaneous speech. Cappa et al. (1994, p. 85) reported a patient with phonemic
jargon aphasia, which they called ‘‘glossolalic jargon’’ whose ‘‘prosodic
quality was appropriate.’’ Moreover, these investigators added that there
were some instances of phonemes present not found in the native language
of the patient (Italian) and also some violations of phonotactic rules ‘‘giving
the listener the impression of a foreign language speaker (‘Spanish’).’’
The anatomical correlates of neologistic jargon in Wernicke’s aphasia
were elaborated by Kertesz and Benson (1970) and Kertesz (1981). In a
series of 10 patients with neologistic jargon, Kertesz and Benson (1970)
consistently found involvement of the dominant posterior superior temporal
gyrus (T1) and arcuate fasciculus. In an additional series of 10 cases, Kertesz
(1981) confirmed that neologistic jargon is most consistently manifested following lesions of the left posterior superior temporal gyrus and the supramarginal gyrus. With regard to phonemic jargon, the bilateral temporoparietal
involvement of the case described by Perecman and Brown (1981) represents
the earliest lesion data previously available on this form of severe jargon.
More recently, Cappa et al. (1994) reported a case of phonemic jargon aphasia (glossolalic jargon) resulting from bilateral temporoparietal lesions.

DISCONNECTED PHONOLOGY

201

In this paper, we report on a patient (J.L.) with dense phonemic jargon
aphasia, who underwent extensive neurolinguistic analysis including phoneme frequency, prosodic features, utterance-final vowel lengthening, and
regional-specific (Southern American English) phonological processes. It is
not uncommon to see intact phonology in patients with impaired speech, but
the point of this paper is to demonstrate the intact nature of the patient’s
phonological organization in the face of virtually complete dissolution of
lexical organization. Our findings suggest that J.L. retained basic phonological processes despite the absence of recognizable English lexical items.
METHOD
Subject. The subject was a 55-year-old right-handed female who was a monolingual native
English speaker. Following sudden onset of dense fluent aphasia, a CT revealed bilateral infarcts, involving the left temporal, frontal, and parietal regions, as well as the right temporal
lobe and posterior frontal area, in the region of the insula (Fig. 1). Following medical stabilization, she was transferred to Baylor Institute for Rehabilitation for intensive inpatient rehabilitation. Upon admission, she underwent standard language assessment, including the Western
Aphasia Battery (WAB) (Kertesz, 1982) and the Boston Naming Test (BNT) (Kaplan, Goodglass, & Weintraub, 1983), as well as the Comprehensive Apraxia Examination (Hanlon &
Brown, 1994). Language assessment included extensive audio and video recordings.
Speech was characterized by nearly total jargon, consisting of fluent strings of phonemes and
intermittent neologisms, with virtually no intelligible utterances (WAB Spontaneous Speech 5
9/20). Narrative speech was characterized by normal rate and prosody, but was intermittently
hypophonic. Not all of her speech was of the same quality. The utterances were variable in
length, some being no more than a single syllable or two, but others having more than 20
syllables. We chose in our analysis to focus on longer stretches of speech estimating that
information about intonation would be easier to extract from these richer structures. Although
she demonstrated occasional perseverative phenomena, as well as normal coarticulation, her
speech was nevertheless analyzable into discrete phonetic segments. As a rough measure of
perseveration, we used the criterion of reduplicated syllable-initial consonants in successive
syllables. Using this approach, she produced approximately ten percent reduplicated syllable
sequences.
Auditory comprehension was grossly defective, characterized by inability to accurately respond to one-step commands or yes/no questions regarding personal information (WAB Auditory Verbal Comprehension 5 16/200). Repetition was markedly defective and completely
jargonized (WAB Repetition 5 0/100). Naming was comparably defective, invariably characterized by phonemic jargon (BNT 5 0/60). Oral reading and reading comprehension were
totally nonfunctional and letter-naming was also defective, with no accurate responses. Written
expression, including copying, writing to dictation and to command, was grossly defective,
consisting of graphemic jargon similar to random letter generation. She demonstrated severe
ideomotor and ideational apraxia, as assessed to command, on imitation, and when attempting
to demonstrate object use. However, she demonstrated no evidence of oral apraxia or apraxia
of speech. She also manifested features of Klüver-Bucy syndrome, specifically hyperorality,
and commonly mouthed and licked objects small enough to be taken to the mouth.
Although her speech was extremely jargonized, it was unmistakable that she spoke Southern
American English. She had the quintessential properties of this variety of American English:
monophthongization of [ai] to [a], vowel breaking of [ε] to [εə], and the lowering and backing
of [ei], cf. Bailey (1985). Pragmatics, including eye contact and conversational turn-taking,
seemed appropriate. She also presented with anosognosia and appeared to have no awareness

202

HANLON AND EDMONDSON

FIGURE 1

DISCONNECTED PHONOLOGY

203

that her speech was incomprehensible. In fact, she accepted jargon produced by the examiner
and readily engaged in jargon conversation.
Procedure. The data for the following computer-assisted, instrumental study of the subject’s
jargonized utterances were taken from the tape recordings made during the language assessment procedures described above. A Sony TCM-5000 Professional Cassette Tape Recorder
provided input to a Speech Box (A-D converter) connected to the parallel interface of a DOS
computer running the CECIL Speech Analysis Software (JAARS International, Waxhaw, NC).
For the purposes of analysis and preservation, a transcription of several assessment sessions
was also undertaken. The transcriptional system chosen was the 1990 version of the International Phonetic Alphabet, according to the conventions of transcription illustrated in Bailey
(1985). A portion of the transcription protocols is provided in Appendix I.
As the subject’s utterances were both rapid and copious and few strings corresponded to
real English words, an interpretation of the raw acoustic signal was often necessary to arrive
at a transcribed form. Instrumental help was called on to assist in making some of these
judgments. For instance, voicing bars seen in the raw wave forms displayed by CECIL were
used to establish the presence of voiced consonants. Also, intensity profiles were employed
as measures of stress to determine the orthotones (the one full-stress per phonological word
phenomenon) in her speech. Voicelessness between vowels helped us to decide the presence
of a two-syllable sequence; pauses, on the other hand (complete lack of sound), were used
to fix word boundaries. The auditory impressions and instrumental appearance of the waveforms for (1) syllable-initial aspiration of stops, (2) syllable-final glottalization of stops, and
(3) Southern American English monophthongization and other regional phonology, were characteristic of her output. We chose not to carry out further quantitative investigation of these
features of the patient’s speech.
Once the segmental aspects of the utterances were determined, a syllable-by-syllable analysis of recorded utterances was completed to divide segmental strings into phonological words.
Also, measurements of suprasegmental features, including pitch and intensity as well as vowel
duration, were taken.
The first task was to determine frequency of occurrence of the subject’s phonemes and to
compare it with the values for American English found in Roberts (1965). The procedure
employed was to count the number of tokens of a particular phoneme (segment), for example,
/n/, in the transcribed section (see Appendix I). The counting of the frequency data required
interpreting the phonetic manifestation of a sound in some cases into its corresponding phonological form (e.g., [é] was interpreted to represent /ə/ for the purposes of this tabulation). This
procedure was repeated for each phoneme that occurred in J.L.’s phonemic inventory, for
example, for /m ə { ŋ t/ and so forth, until all phonemes had been counted. The number of
tokens of a particular phonological segment in J.L.’s speech was then divided by the total
number of segments produced in that utterance to determine the percentage of occurrence. The
percentage of occurrence for each phoneme in J.L.’s speech was compared to the percentage of
occurrence found by Roberts (1965) and plotted with J.L.’s values on the X axis and the values
for normal spoken English on the Y axis. (The system of phonemic notation used by Roberts
is that of Trager & Smith (1958), which we have altered in our notation to reflect current IPA
usage, e.g., /j/ not /y/. Roberts’ also employed the following segmental inventory: Vowels /
i u e ə o a { ɔ/ and consonants /p t k b d g T D f ɵ v ð s ʃ z Z m n ŋ l j h r w/.)
We could easily observe in the patient’s speech periodic elevations of the pitch followed
by gradual lowering and then a longer pause, cf. Fig. 3. We took these long-term characteristic
prosodic profiles as initial indications of the phonological phrase. Ten simple declarative
phrases so determined were selected from J.L.’s corpus of utterances for quantitative analysis.
These were digitized using the CECIL version 2.0. Hardware and Software Speech Analysis
System. CECIL’s pitch-extraction algorithms displayed the fundamental frequency as a track
across the screen of the CRT. Programmable cursors were placed at the onset of voicing of
the first syllable of the utterance and also at the offset of the last syllable. F∅ values were
taken for these two points as well as for duration and fundamental frequency declinations.
The same ten utterances were used to determine the distance between orthotones (stressed

204

HANLON AND EDMONDSON

vowel syllables in a phonological word). The digitized utterance (raw wave form) was placed
in one window at the top of the CRT and the intensity was displayed in a window below it.
It was visually evident that there were a number of intensity prominences followed by reduced
peaks at intervals across the utterance. The programmable cursors were used to measure the
duration between these high points.
Duration of phrase final stressed vowel length was measured. According to Lindbloom &
Rapp (1973), Cooper, LaPointe, and Paccia (1977), and Sorensen and Cooper (1980), for both
English and Swedish, normal speakers demonstrate vowel lengthening in their utterances;
duration of stressed vowels is longest in phrase final (Vf ) position, as compared to the durations
at phrase initial (Vi ) and phrase medial (Vm ) positions. The same ten utterances used to determine J.L.’s fundamental frequency declination values were also used to study phrase final
lengthening effects. The vowel duration of the first syllable in each of the ten was measured
from the first vowel burst at syllable onset. Similarly, a medial vowel duration was measured
at a point approximately halfway through the utterance. And finally, the vowel duration of
the last syllable in each utterance was established.

RESULTS

Table 1 and Fig. 2 compare the subject’s phoneme frequency with the
normative data on phoneme frequency found by Roberts (1965) for spoken
American English. Table 1 lists the normative frequency of the American
English segments and comparable values for the patient. Roberts (1965) determined that /ə/ occurred most frequently (11.82 percent) followed in turn
by /i t j r n e a w s l d h m k ð z/ ranging from 9.29 to 2.0 percent. The
remaining English phonemes occurred much less frequently. The corresponding phoneme frequency of the subject was surprisingly similar to normative data. Only one phoneme /ə/ occurred more than 10 percent of the
time. Six segments occurred more than 5 percent of the time. And the remainder occurred less frequently. In other words, the frequency distribution does
not demonstrate values that radically differ from the norm.
Figure 2 presents the normative segment frequency in comparison to that
of the subject. We found a significant correlation between the normative data
and J.L.’s segment frequency r 5 1.76, p , .001). This comparison resulted
in a regression line with a slope of .79 and an intercept of .008 with 58%
of the variance accounted for by this line.
Table 2 presents F∅ at onset, F∅ at offset, length of utterance, number
of vowels, duration of initial vowels, duration of medial vowels, and duration
of final vowels. A representative utterance is exhibited in Fig 3. The measurement of ten utterances was used to establish the ‘‘top line.’’ By measuring
the fundamental frequency at the onset of a polysyllabic utterance, the mean
value found was 42.0 6 1.6 semitones (the relationship between semitones
and the more familiar Hertz scale is fST 5 39.86 3 log10 (fHZ /16.35), cf.
Ross, Edmondson, & Seibert (1986) for discussion of the hertz vs. semitones
scale). As can be seen in Table 2, the onset frequency was rather tightly
clustered around the mean with a range of 39.8 semitones in utterance #2
to 44.3 semitones in utterance #7. (The particular utterances used in this task

DISCONNECTED PHONOLOGY

205

TABLE 1
Comparison of Phoneme Frequency
Phoneme

J. L.

Normal English
speakers

ʃ
Z
ŋ
ð
ɔ
dz
p
o
θ
v
h
z
j
w
b
g
u
d
e
m
f
k
s
r
t
l
{
i
a
n
ə

0
0
.0053
.0057
.007
.0088
.0088
.0088
.0105
.0105
.0123
.0123
.0123
.0158
.0176
.0193
.0228
.0333
.0351
.0422
.0439
.0439
.0457
.0492
.0509
.0545
.0562
.058
.0615
.0756
.1195

.0072
.0003
.0092
.0225
.0064
.0036
.0161
.0154
.0042
.0188
.0263
.02
.0677
.0452
.0163
.0087
.0191
.0304
.0474
.0261
.017
.0245
.039
.0658
.0695
.033
.0154
.0929
.0463
.0629
.1182

are indicated by the number in parentheses in the transcribed protocols given
in Appendix I.)
Measurement of the fundamental frequency of offset was also completed
and a mean value of 36.2 6 2.9 semitones was determined. There was a
somewhat looser distribution of offset values than of onset values. The maximum occurred in utterances #3, #5, and #6 and was equal to 39.6 semitones,
whereas the minimum occurred in utterance #1 and was equal to 31.7 semitones.
The values of F∅ declination of J.L. were compared to the values found
by Cooper & Sorensen (1981) in their study of fundamental frequency in
sentence production. They determined that the mean frequency of ten speak-

206

HANLON AND EDMONDSON

FIG. 2. Scatterplot of subject’s vs. normals’ segment frequency.

ers was 200 Hz in the initial stressed syllable and 159 Hz in the final syllable
of test sentences. If one converts these values to semitones to allow a comparison to the patient J.L., then Cooper & Sorensen’s normals had values of
43.35 semitones at onset and 39.38 semitones offset, whereas J.L. had the
values 42.01 at onset and 36.20 at offset, cf. Table 2.
Duration was determined for each utterance by calculating the time of
onset minus the time of offset of voicing, resulting in a mean value of 2.18
6 0.80 secs. The length of utterance values varied considerably with a range
of 1.2 secs in utterance #4 to 3.5 secs in utterance #10. The duration of the
utterance directly reflected the number of vowels, whereby the greater the
number of vowels, the longer the duration of the utterance, generally.
TABLE 2
Frequency Declination and Utterance-Final Vowel Lengthening
Onset
Offset
Length of Utt
Length Initial Length Medial Length Final
Utterance (freq in st) (freq in st)
(sec)
# of Vls
(sec)
(sec)
(sec)
1
2
3
4
5
6
7
8
9
10
Mean
SD

40.10
39.80
45.00
40.70
42.30
41.00
44.30
43.00
42.80
41.10
42.01
1.67

31.70
38.80
39.60
36.00
39.60
39.60
34.90
32.30
34.00
35.50
36.20
2.90

2.25
3.03
1.41
1.23
1.87
1.99
1.90
3.36
1.28
3.50
2.18
0.80

6
7
4
3
7
5
5
6
3
6
5.2
1.4

0.31
0.18
0.18
0.17
0.17
0.20
0.35
0.20
0.13
0.18
0.21
0.07

0.20
0.25
0.24
0.22
0.17
0.30
0.30
0.22
0.27
0.25
0.24
0.04

0.21
0.39
0.44
0.24
0.38
0.53
0.47
0.60
0.23
0.26
0.37
0.13

DISCONNECTED PHONOLOGY

207

FIG. 3. Example of pitch declination.

The utterance with the greatest number of syllables (i.e., 7 vowels) had a
duration of 3.0 secs, whereas the trisyllabic utterance (#9) lasted only 1.3
secs. The correspondence between syllable number and total duration, however, was not exact. Utterance #5, for example, contained 7 vowels, but had
a duration of only 1.9 secs.
We also completed an analysis of vowel lengthening in utterance-final
position, as compared to vowel length in utterance-initial and utterancemedial positions. As can be seen in Table 2, as well, the vowels in initial
position in the 10 utterances had a mean duration of .21 6 .07 secs. Those
in medial position had a duration of .24 6 .04 secs, and those in utterancefinal position demonstrated a duration of .37 6 .13 secs. The vowels obviously increased in duration as a function of utterance position with utteranceinitial being the shortest, utterance-final the longest, and utterance-medial
position occupying a middle value.
The findings for the patient were consistent with the values of utterancefinal vowel lengthening determined by Oller (1973) and Cooper & Sorensen
(1981, p. 32–33) for normal English speakers. They used a slightly different
procedure in which the initial syllable was set to the unit length and the times
of the medial and final syllables were then determined as a factor of the
duration of the initial syllable. Converting the values of J.L. given in Table
2 in the same fashion shows the following result. Normals in initial syllables
had a value of 1.00 6 0, in medial syllables of 1.52 6 .07, and in final
syllables of 2.12 6 .13, whereas the patient had values in the initial syllable
of 1.00 6 0, in the medial syllable of 1.25 6 .37, and in the final syllable
of 1.90 6 .56.

208

HANLON AND EDMONDSON

DISCUSSION

The present case represents an example of pure phonemic jargon, in the
sense that J.L. produced fluent phonologically intact utterances that contained virtually no lexical items of English. Similar to the previous reports
of patients with phonemic jargon aphasia by Perecman and Brown (1981)
and Cappa et al. (1994), the language dissolution of the present case was
secondary to bihemispheric lesions. All three of these cases of phonemic
jargon involved bilateral temporal infarcts, suggesting that involvement of
the right temporal area may be necessary to produce a language impairment
of this magnitude.
Despite her jargonized speech, our investigation showed a relatively intact
inventory of phonological segments and that these segments occurred with
a frequency not unlike that found in normal speech. Moreover, we found no
clear cases of non-English phonemes in her utterances nor examples of obvious violations of English phonotactics, unlike the Cappa et al. (1994) case
mentioned above. The phonetic-to-phoneme linkage expressed through the
idea of phonological processes appeared to be undisturbed, as we concluded
from the fact that her phonetic units were very much like the units of normal
speakers and showed the distribution and change in context as found in English. What was impaired was that she no longer had access to the lexical
representation of English words.
Not only was her segmental phonology relatively undisturbed, her longer
term prosodic phonology also resembled normal speech. For example, she
demonstrated the ‘‘top line’’ rule with concomitant F∅ declination and
stressed vowel lengthening utterance-final, characteristic of spoken English.
Beyond that, her speech had many phonological characteristics of Southern
States American English, including monophthongization, (e.g., [skai] sky
becomes [ska]), vowel breaking, (e.g., wεË] well becomes [wεəË], and lowering and backing of [ei], (e.g., [wεit] weight becomes [wəit].
English has been classified as a stress-timed language (Bailey, 1985), in
which each stressed syllable is separated by about the same amount of time,
unlike French or Spanish, which are syllable-timed languages, in which each
syllable occupies about the same amount of time. Moreover, each phonological word in English contains a single syllable with main stress, an orthotone.
This characteristic feature of English prosody appeared undisturbed in the
speech of J.L. Measurements of vowel duration, intensity, and pitch all
showed that J.L. retained the orthotonic rhythm of the phonological sentence
that is characteristic of English speech. (See Anderson, 1981.)
Consistent with the findings of Cooper, Danly, and Hamby (1979) and
Cooper and Sorensen (1981), J.L.’s F∅ declination exhibited most of the
features found in normal speakers of English. Secondly, although she demonstrated the phonetic segments that correspond to English phonemes, she
failed to demonstrate the organization of the phonemes into the lexical words

DISCONNECTED PHONOLOGY

209

of the language, resulting in jargonized output more severely disturbed than
the classic paraphasic and neologistic errors of Wernicke’s aphasics. In summary, this case of phonemic jargon aphasia provides evidence that the basic
prosodic features of language, basic processes of phonology, and even some
properties of regional vernacular can be preserved while lexically driven
phoneme choice is impaired. That is to say, this neurolinguistic analysis suggests phonology can be disconnected from the lexical inventory of forms
that make up the material of words so that phonological words can be intact
while lexical words can be nonexistent.
APPENDIX I
Transcription of J.L.’s Speech

A 5 therapist
B 5 patient
Dialogue 1:
A: Tell me how you slept last night. OK?
B: 'h{ʔε 'Dεlən 'thr{ʔnµ 'thr{k 'nəznµ (1)
A: Uh huh!
B: 'θei 'gɑdə 'h{dé z{n 'grau ət nµ 'f{s (2)
A: Uh huh! Was somebody loud?
B: [Yeah.] 'ε̃ʔ 'lεhεm 'daəm 'thaəm. [Yeah]. (3)
A: Uh huh!
B: [Ya know]. jə 'Də 'thεl θə 'f{ks 'rae t. (4)
A: OK!
B: 'kh{ən bə'lεʔµl'lεʔµl 'laəs 'saəd.
A: So, What did you do about it? What?
Did you do anything?
B: 's2
A: Tell me.
˜ 'hw{Du 'f{ "Du'w { ni 'kh{nikh{n"kh{n. (5)
B: 'hw{
A: Uh huh!
B: "nou'f{s tə də'liə 'khin 'gεiəm. (6)
A: Is that all?
B: 'bin 'fizl.
'ðs 'nauə, 'sə 'flf əvə 'fs 'f{ndə 'mεi 'zŋ skhεim ə'kh l{mə'khauə. (7)
{n khə'rɑmən 'phlεis ə 'thrvéʔ o2 'ʔ 'khrε̃m2 'bεi. (8)
A: Uh huh. Are you still talking about last night?
B: [Yeah.]
A: OK. Go on.
B: [Well.] 'a də'dεs. [I don’t] 'vɑməv.
'walə'walə 'wanə thə'liŋ 'w2 bj2ʔ. (9)

210

HANLON AND EDMONDSON

'gεs hiz dεəʔ 'thεinén 'khεinén.
A: Uh huh.
B: (Inaudible).
A: So why were you in such a good mood when I came up to get you?
B: (Hypophonic speech.)
A: Speak up. I can’t hear you.
(Long pause.)
B: 'nεmə 'mai 'məŋ 'spəŋ '{mə 'sənəp mə'bəis 'səi mə'seəlf. (Pause). 'hwɑdəjə
'nou ae gə'g2θ phə. ga'lani 'galni (Pause.) (Inaudible)
'sou jə 'gaʔ də'brε f{əs é'nəki'nək. (Long pause.) feəl pheəl. haifé'n{k 'sf. (Long
pause.) 'Dju 'h{l mi? (10)
Dialogue 2:
A: How are you today?
B: Hypophonic speech.
A: How are you today?
B: 'hauʔ ju meik mi. 'rɑni 'gεtnµ 'glεtµn in'mauʔ 'məʔn min'au 'minə 'khenʔ θeiθ2
dədei dei'dŋ 'thei. (Pause)
'ou 'hw{dai 'es khə'vεk 'thi (Inaudible).
A: OK. Hey! Have you been in here before?
B: [Today?]
A: Uh. Huh.
Have you been in here before, today?
B: Uh. Huh. (Pause).
A: What is your name?
B: [Well.] (Pause) 'reəl 'bɑlb 'b{lmöz 'bötsi {z 'riəl. 's{m 'fou 'fɑ2 'fe. {n
'khau fs 's{mə 'leik 'feida feij{'vεk anmə'vεk. 'fεtə 'f{k.
A: Hm. OK. What is your address?
B: Hypophonic speech.
A: Speak up! Hm! OK.
Where do you live?
B: [I don’t even] 'dau '{k 'th rei 'd{2 'r2 'rε2t 'phrauən 'frəm εim 'frəmən.
(Pause).
ɑlə'ðəðə 'lɔθµl ɑl'stəfə'nəʔ. 'h{və"voulə εm'bɑdəs 'ʔet 'θei 'nöəsnµ . (Pause.)
'θeiəsən. ai'lε 'fö 'fεθ 'fö 'feiən 'thiŋ 't haiə 'mε2 'hε2 'sku [because]pə'khöz [he
is] 'heiʔəz 'mɑ"di '{lraits .
Example of a response to a Yes/No question from WAB:
A: Are you wearing red pajamas?
B: nµ 'h{səm 'ɑlés 'gidə 'θεsə 'ʔɑ 'ʔɑlə 'sprau 'mauθ khɑ2 'Dögə ʔə'tɑ2mDə
'skɑhɑ 'dəŋ məkhεtʔ bə'lɑdə"lɑdə 'ʔ2dən. 'hno 'rɑnʔ (Pause) 'flnµl 'θiŋ
'jölə"n2.

DISCONNECTED PHONOLOGY

211

Examples of responses to BNT items:
A: What is that called? (Item no. 1 ‘‘bed’’)
B: thɑ 'fɑ fə
A: What’s that called? (Item no. 2 ‘‘tree’’)
B: 'ö mə'lɔŋ oi'jεə mɔ'jau. maumo2 'thai.
A: Alright. Let’s try this next one. (Item no. 3 ‘‘pencil’’)
What is this called?
B: (Inaudible)
A: OK. What’s that? (Item no. 4 ‘‘house’’)
B: 'tʃ h{thə
A: Alright. The name of this thing starts with? [hɑ]
B: tsei'ʔai2.
A: What is this called? (Item no. 5 ‘‘whistle’’)
B: No response.
A: What do you do with this thing? (Item no. 5 ‘‘whistle’’)
˜ non 'doi 'w{k 'mɔst də'z 'ʔoumi 'h{d 'kht 'bi 'gd"n 'm{t. 'wɔ 'nusə.
B: h{
(Laughter) 'hi wənə'wədə. (Laughter) 'wɑthə
A: Alright. The name of this thing starts with? ,w. (‘‘whistle’’)
B: ŋ{. nɑ'd{də'l{t bou'h{ də'l{t gə'khei'khein.

REFERENCES
Alajouanine, T. (1956). Verbal realization in aphasia. Brain, 79, 1—28.
Anderson, S. W. (1981). Vowel timing and linguistic organization of articulatory sequences in
jargonaphasia. In J. W. Brown (Ed.), Jargonaphasia (pp. 295–304). New York: Academic
Press.
Bailey, C-J. (1985). English phonetic transcription. Dallas: UTA/SIL Series in Linguistics.
Buckingham, H. (1987). Phonemic paraphasias and psycholinguistic production models for
neologistic jargon. Aphasiology, 1, 381–400.
Buckingham, H., Avakian-Whitaker, H., & Whitaker, H. (1978). Alliteration and assonance
in neologistic jargon aphasia. Cortex, 14, 365–380.
Buckingham H., & Kertesz, A. (1974). Neologistic jargon aphasia. Amsterdam: Swets &
Zeitlinger.
Butterworth, B. (1979). Hesitation and the production of verbal paraphasias and neologisms
in jargon aphasia. Brain and Language, 8, 133–161.
Cappa, S. F., Miozzo, A., & Frugoni, M. (1994). Glossolalic jargon after a right hemispheric
stroke in a patient with Wernicke’s aphasia. Aphasiology, 8, 83–87.
Cooper, W. E., LaPointe, S. G., & Paccia, J. M. (1977). Syntactic blocking of phonological
rules in speech production. Journal of the Acoustical Society of America, 61, 1314–1320.
Cooper, W. E., Danly, M., & Hamby, S. (1979). Fundamental frequency (F∅) attributes in
the speech of Wernicke’s aphasics. In J. J. Wolf & D. H. Klatt (Eds.), Speech communication papers presented at the 97th meeting of the Acoustical Society of America. New
York: Acoustical Society of America.
Cooper, W. E., & Sorensen, J. M. (1981). Fundamental frequency in sentence production.
New York: Springer-Verlag.

212

HANLON AND EDMONDSON

Denes, P. B. (1963). On the statistics of spoken English. Journal of the Acoustical Society of
America, 35, 892–905.
Hanlon, R. E., & Brown, J. W. (1994). Apraxia. In V. S. Ramachandran (Ed.), Encyclopedia
of human behavior. San Diego: Academic Press.
Kaplan, E., Goodglass, H., & Weintraub, S. (1983). Boston Naming Test. Philadelphia: Lea &
Febiger.
Kertesz, A. (1981). The anatomy of jargon. In J. W. Brown (Ed.), Jargonaphasia (pp. 63–
112). New York: Academic Press.
Kertesz, A. (1982). Western Aphasia Battery. Orlando: Grune & Stratton.
Kertesz, A., & Benson, D. F. (1970). Neologistic jargon: A clinicopathological study. Cortex,
6, 362–386.
Kotten, A. (1982). Therapy of neologistic jargon aphasia: A case report. British Journal of
Disorders of Communication, 17, 61–63.
Lebrun, Y. (1987). Anosognosia in aphasics. Cortex, 23, 251–263.
Lecours, A. R., & Rouillon, F. (1976). Neurolinguistic analysis of jargonaphasia and jargonagraphia. In H. Whitaker & H. Whitaker (Eds.), Studies in neurolinguistics (vol. 2) (pp.
95–144). New York: Academic Press.
Lhermitte, F., Lecours, A. R., Ducarne, B., & Escourolla, R. (1973). Unexpected anatomical
findings in a case of fluent jargon aphasia. Cortex, 9, 436–449.
Lindbloom, B., & Rapp, K. (1973). Some temporal regularities of spoken Swedish. Papers
from the Institute of Linguistics (University of Stockholm), Volume 21, 1–58.
O’Connell, P. F. (1981). Neologistic jargon aphasia: A case report. Brain and Language, 12,
292–302.
Oller, D. K. 1973. The effect of position in utterance on speech segment duration in English.
Journal of the Acoustical Society of America, 54, 1235–1247.
Perecman, E., & Brown, J. W. (1981). Phonemic jargon: A case report. In J. W. Brown (Ed.),
Jargonaphasia (pp. 177–257). New York: Academic Press.
Perecman, E., & Brown, J. W. (1985). Ukeles, condessors, and fosetch. Language Sciences,
7, 177–214.
Roberts, A. H. (1965). A statistical linguistic analysis of American English. The Hague:
Mouton.
Ross, E. D., Edmondson, J., & Seibert, G. B. (1986). Effect of affect on various acoustic
measures of prosody in tone and non-tone languages: A comparison based on computer
analysis of voice. Journal of Phonetics, 14, 283–302.
Semenza, C., Cipolotti, L., & Denes, G. (1992). Reading aloud in jargonaphasia: An unusual
dissociation in speech output. Journal of Neurology, Neurosurgery, and Psychiatry, 55,
205–208.
Sorensen, J. M., & Cooper, W. E. (1980). Syntactic coding of fundamental frequency in speech
production. In R. A. Cole (Ed.), Perception and production of fluent speech (pp. 399–
440). Hillsdale, NJ: Erlbaum.
Trager, G. L., & Smith, H. L. (1958). The structure of American English. New York: The
Ronald Press Co.
Weinstein, E. A., Lyerly, O. G., Cole, M., & Ozer, M. N. (1966). Meaning in jargon asphasia.
Cortex, 2, 165–187.
Weinstein, E. A., & Lyerly, O. G. (1976). Personality factors in jargon asphasia. Cortex, 12,
122–133.