CTLF Corpus de textes linguistiques fondamentaux • IMPRIMER • RETOUR ÉCRAN
CTLF - Menu général - Textes

Stetson, Raymond. Motor Phonetics – T05

Appendix IX
American version of phonemics (phonologie)

The following may be cited as representative of the
American version of Trubetzkoy's Phonologie:

Trager, G. L., & Bloch, B. Syllabic phonemes of English.
Lang., 1941, 17. TB, (97)

Bloch, B., & Trager, G. L. Outline of linguistic analysis.
Ling. Soc. Am. 1942. BT, (6)

Hockett, C. F. System of descriptive phonology.
Lang., 1942, 18. HS, (31)

Harris, Z. S. Phonemicist version of Yokuts.
IJAL, 1944, 10. HY, (24)

— Phonemicist version of Navaho.
IJAL, 1945, 11. HN, (25)

— Phonemicist version of Kota.
IJAL, 1945, 11. HK, (26)


There is no clear statement of the articulatory
nature of the phoneme. The definition is allowed
to stand in the logistic terms of Sweet-Saussure-Trubetzkoy:
a phonic difference which distinguishes
one word from another constitutes a phonemic difference.
Often the phoneme is treated as a differential;
and often it is counted a class of articulations, or
of articulatory attributes. The phonic difference must
be due to a controlled movement (muscular contraction)
which modifies the articulation. It is a class of
articulations in that a series of articulations may be
modified by this same muscular contraction. The
nature of the “opposition”, whether a group of
related shapes, as with the vowels, or as the presence
and absence of a particular attribute, is determined
by the apparatus in terms of which the phonic
differences and therefore the definition of the phoneme
must be formulated.

In motor phonetic terms the phoneme is a characterized
factor of a syllable. The factors of the
syllable are the canal shape (the vowel through
which the syllable pulse is emitted), and the release
of the pulse by the chest muscles or by consonant,
and the arrest of the pulse by the chest muscles or
by consonant. Various aspects of the syllable factor
characterize the syllable: vocalization, nasalization,
constriction, etc. The phoneme always involves a
syllable factor. The classes to which the phonemes
are reduced are classes of syllables produced by
these characterized syllable factors. The simplest
“word” to be differentiated by a phoneme is a
syllable. The phoneme can appear only in a syllable.

Formation of Speech Sounds: BT (6, p. 12)

The statements made as to the production of
human speech are quite wrong, as experimental
study of the speech apparatus shows. A free flow
of a column of air under steady pressure through an
enclosed passage is postulated, and the mechanism
is compared to a clarinet or flute. It is “a kind of
‘playing’ on the column of air” which is said to
produce the sounds of human speech. The notion of
a free, steady flow, is false; the air is emitted in
pulses, and the chest pressure between syllables is
often neutral. The clarinet and flute are not analogous
to speech; the comparison should be with the
brass of the orchestra, the notes of which are made
by pulses of air from the chest, precisely as the
syllables are made by the speech apparatus. The
study of speech with an artificial larynx reveals the
function of the chest in making the individual
syllable pulses. The reed of the artificial larynx
is always ready to sound; the stopping and starting
of the syllables must be done by the chest muscles.
The function of the releasing and of the arresting
consonants and the whole structure of the syllable
cannot be properly understood unless the fundamental
movement of the syllable pulse is clearly conceived
(87; 91; 64, p. 319).166

Simultaneous Components of the Phoneme: HN (25,
p. 243).

On logistic grounds, Harris proposes to break up the
phonemes into simultaneous components. No articulatory
mechanism is proposed, and the analyses
are not consistent. ; T and ʼ are assumed to give a
single phoneme ; a glottalized t is quite possible;
but h and l are assumed to give l, in which the
aspiration of the h cannot coexist with the voiced
l in t.

In motor phonetic terms, such a simultaneous
component will be defined as a significant accessory
movement concurrent with the syllable factor: i.e.
glottalization, vocalization, nasalization, etc.

Sound Types: TB (97, p. 223, 223 n)

“The sound-types constituting a phoneme must be
phonetically similar, complementarily distributed,
and congruently patterned; and the class thus composed
must be in contrast with and mutually exclusive
of every other such class in the language.
Sound-types, as members of a phonemic class, are
called allophones. The sound-type is a class of
phonetic events called sounds; each sound is a sum
of sound-features (as voicing, aspiration, occlusion,
labial position, etc.) which may occur in various
combinations. The repetition of what is perceptually,
the same combination constitutes the sound-type,
which is thus an abstraction from a series of
utterances clustering about a norm.”

In motor phonetic terms, “perceptually the same
combination” means conditioned to meaning as the
same articulation, in the context in which it occurs…
in which it has been conditioned. Therefore
the phoneme is not “a series of utterances
clustering about a norm” because the conditioning
does not concern a series of minor variations but
concerns an articulation with variants so different
that they may coincide with the variants of other
phonemes; and may be reduced to zero. The context
is all important in determining the conditioned
response. The users of the language have come to
recognize the variant of the phoneme in that context.
The use of the term “type” is good; the
conditioned articulation is the “type” with an
indefinite number of variants in the varying stimulus
pattern to which the movement is conditioned. In
cases where there is overlapping of the variants of
different phonemes, it is easy to identify the phoneme
by changing the rate and stress.

Phoneme as Segmental: BT (6, p. 41)

“May be regarded as segments of utterances.” Such
a definition of “segment” is quite inadequate. They
are said to follow each other in the stream of speech,
but the syllable factors are not merely sequential;
they inevitably overlap (64). The fundamental
representation of “sounds following each other”
and therefore of “segments” corresponding to the
“sounds”, is at fault. The notion of playing sounds
like the notes on a flute or clarinet was mistaken
to start with.

Allophones: HS (31, p. 9)

Variants of a given “sound” are obvious. The
context determines the variants. The varying function
of the consonants in the syllable, the compounding
of consonants, and the influence of stress and
grouping on the vowels all modify the type movement
of the syllable factor. To account for these
variants within the phoneme, the term “allophone”
has been devised.

In motor phonetic terms such variants are
referred to the varying functions of the syllable
factors, and the reciprocal influence of the movements.

Morphophonemes: HN (25, p. 243, 245), HK (26,
p. 285)

In discussing Emeneau's treatment of Kota, Z. S.
Harris speaks of the reasons for assuming morphophonemes.
“For every morphoneme he assumes a
base form, which is composed not of phonemes but
of morphophonemes. When the morphonemes occur
in words (i.e. in speech), the morphophonemes
of their base forms are replaced by the corresponding
phonemes: morphoneme k by phoneme /k/, and
so on. In some cases the replacement is not by the
167corresponding phoneme but by some other; e.g.
morphophoneme n, when it is due to follow a
morphophoneme l, is replaced by phoneme /n/.
This occasional non-corresponding replacement is,
of course, the only reason for the setting up of base
forms and morphophonemes.” This is the clumsy
result of setting up logistic units, instead of dealing
with movements which inevitably influence each
other. But the Kota system has to be carried out
with further complications, and there prove to be
“two levels of replacement”, etc. The morphophoneme
is a device for logistic classification of
phonemic variants directly due to context. The
logistic method does not provide for a changing
articulation in context, but names each stage of the
change as a separate entity, the “morphophoneme”,
and then collates them with the appropriate phonemes.

Instead of such cumbersome morphophonemes
replaced by phonemes, etc., motor phonetics deals
with the changes in the type form n when it follows
l; experimental observation will show that the
change is due to the influence of one movement on
another and this makes all the logistic paraphernalia

Prosodic Features: Quantity, Accent, Rime, etc.:
TB (97,p. 238),BT (6, p. 34)

Suprasegmental Features: TB (97, p. 224), BT (6,
p. 41, 49-50), HS (31, p. 8), HY (24, p. 205)

These various terms are applied to the traits of
the syllables in feet and in breath groups. The
term “suprasegmental” is general, and might apply
to any and all of the “prosodic features”. The
prosodic features refer to the rhythmic organization
of the feet and breath groups. But in all cases the
analysis is based on a string of phonemes to which
these features are added; there is no means of organizing
the prosodic features and the phonemes
into a common movement, and yet the practical
handling of such features always involves the unit
of the syllable. The logistic scheme adds terms but
provides no organization.

In motor phonetic terms the syllable proves the
fundamental pulse in which the phonemes have
their function as syllable factors, and the syllable
pulses are organized by larger inclusive movements
into the feet, and the breath groups. It is a matter
of the coordination of movements; the syllable pulses
are produced by the chest muscles; they become a
part of the coordinating inclusive movements of the
feet and breath groups produced by the abdominal
movements involving muscles of abdomen-diaphragm.

The Vowel — Consonant Contrast: BT (6, p. 18)

This is a fundamental contrast, and appears in all
phonological treatments. The contrast depends on
the radically different functions of the consonants,
which delimit the syllable pulse, and the vowel,
which shapes the canal to emit the pulse. But the
logistic handling does not recognize the syllable in
which the consonants and vowels function, but
deals in “position” of phonemes in reference to each
other and in “oppositions” of phonemes. So the
distinction, though recognized, is said to be “blurred”

In motor phonetic terms the syllable is the pulse
in which the consonants and vowels are factors; the
function of the consonant and of the vowel in the
syllable is always distinct; and there is no chance of
“blurring” the distinction.

Clusters, Doubles: TB (97, p. 229), HN (25, p. 240-242),
HS (31, p. 10)

Since the logistic system does not specify a function
for the consonants and vowels, the “cluster” remains
an arbitrary sequence of two or more consonants,
or of two vowels. The phonemicists are consistent
in making clusters of vowels as well as of consonants,
although the combination is quite different.

Motor phonetics deals with the clusters of consonants
as compound consonants and abutting consonants;
and with clusters of vowels as diphthongs.
The releasing or arresting compound results from
a fusion of the movement of two or three consonants
into a single consonant stroke; and it
168can be shown that there are conditions in which
the compound consonants will become abutting
consonants, and vice versa. The diphthongs result
from the fusion of two vowel shapes; but unlike
the consonants, the movements are consecutive —
the shape changes during the syllable; it is usually
possible to show how the diphthong developed.

Distribution, Structure: BT (6, p. 45 f.)

Distribution of phonemes is the logistic equivalent
of the system of habits which make up the language
mechanism conditioning the occurrence of the phonemes.
The logistic analysis undertakes to handle
the structure by noting the permissible sequence and
contrasts. The classification of the phonemes results.

This amounts to the classification of the motorphonetic
syllable and of its factors; and the classification
of the function of the syllable in the foot
and breath group.

Syllable Appears Unannounced: TB (97, p. 224),
HY (24, p. 206), BT (6, p. 22)

Since the phoneme is the primary unit, and is
defined without reference to the syllable, the syllable
is treated as ‘suprasegmental’, as something
added to the phoneme series. No provision is made
for the sylllable in the original specifications. Hence
it appears as an arbitrary addition to the phoneme

In some cases the syllable is avoided, as in Z. S.
Harris' version of Yokuts. As it is not counted
essential, a formula which ignores the syllabic seems
more general.

The definition of the syllable reverts to “sonority”
(or “prominence”) although the concept has no
place in acoustics, and is meaningless. The syllable
is obvious, as they all agree, and it is natural to use
a paraphrase of that fact, and count the distinguishing
trait “prominence”.

The motor phonetic definition of the syllable
makes it the basic chest pulse; the phonemes have
their occurrence in the syllable only; it is an inclusive
movement in which the syllable factors function,
not a “suprasegmental” unit added to the phoneme

Vowels Nuclear, Consonants Marginal — TB (97,
p. 224)

Syllable Implied — HS (31, p. 10, 14)

Base Forms —HK (26, p. 284-285), HN (25,
p. 245)

Syllabic Consonants — BT (6, p. 28)

Coarticulation — BT (6, p. 29)

Syllabic Division; ambisyllabic Consonants — TB
(97, p. 234)

Onset of Stress —TB (97, p. 234)

All these terms fall into line when the syllable
is made fundamental; they consist of various generalized
observations of the phases of the syllable.

Phonemic Phrases — Stress Culmination — TB
(97, p. 226), HS (31, p. 6)

Contour of Stress — BT (6, p. 41)

Intonation — BT (6, p. 42)

This is the term of the phonemicists for the
breath group. In one form or another, the breath
group is recognized, since stress and grouping play
an important part in morphology and syntax.

In motor phonetics the breath group appears as
the inclusive movement for the syllables and feet;
it is an inevitable coordination of the breathing
movements in speech.

Junctures and Boundary Markers: TB (97, p. 225)

Pause as a unit: TB (97, p. 225)

Juncture as a gap or as an entity: HN (25, p. 240,
243, 244)

The joining of larger units of speech forces attention
because the phonemes are modified at the
joining of these larger units. Although the phonemicists
ignore the inclusive movements, they are
aware that the boundary markers at the ends of
the larger units are like the boundary markers
between subordinate units, and therefore generalize
the concept of juncture to cover all boundary
phenomena. It would be better to count a juncture
169the case where two termini come together, where
two ends abut. This would have eliminated the
inconsistency of an “external open juncture” which
does not join anything.

In motor phonetic terms, the beginning and the
end of the inclusive movements are naturally marked,
and the syllables and syllable factors which
come together between movements are modified
by the juxtaposition, which can be studied in its
detail in the laboratory.

External Open Juncture: HK (26, p. 283 f.)

Internal Open Juncture:

Close juncture:

These junctures are actually terminal phenomena.
They become “junctures” only in the case
of “internal open juncture” and “close juncture”
between breath groups, feet, and syllables. Within
the syllable, conditions are entirely different, although
the phonemicists have not taken the difference
into account. The logistic handling is such
that internal open juncture and close juncture may
alternate, depending on the rate and stress. The
observations are not consistent, and there is no
consideration of the variations with changes of rate
and of stress.

The boundary markers (Grenzsignale) define the
inclusive movements. They assume importance in
a logistic system because they are the only recognition
of the coordination of the movements which
underlies the process of speech.

In a motor phonetics the boundary marks have
their natural place as the indications of the limits
of the movements which folow each other, but
which are knit together by the underlying movement.170