Background

1

Feet as Neuropsychologically Driven Sense
Units Fusa Katada Waseda University
Abstract
This paper seeks a way to define feet as a
substantive universal, just like phonemes,
minimal speech units, as a substantive universal
arising naturally from the physiological
structure of our articulatory organs. It
proposes that all languages be looked at having
feet as minimal perceptual/productive sense units
(PSU in the sense of Kohno 1993/Uemura 1997) that
arise naturally from the neuropsychological/physi
ological structure. It is shown that the PSU
foot is an isochronous unit which underlies
three types of languages classified in terms of
rhythm stress-timed (Abercrombie 1967),
syllable-timed, and mora-timed.
Background Since the work on English stress by
Liberman (1975) and Liberman Prince (1977), the
metrical foot has been studied largely as
synonymous with the stress foot, a fundamental
isochronous unit of stress-timed languages such
as English. The foot in this sense, then, should
not be a relevant notion for non-stress-timed
languages that do not utilize stress, such as the
mora-timed language Japanese. However, a higher
metrical unit controlling two moras has been
established in the language and this unit is
referred to as bimoraic foot (Poser 1990).
Fundamental questions then arise (1) Are
stress foot and bimoraic foot two complementarily
shared units between stress-timed
languages and mora-timed languages? (2) Or,
would they ever be substantive universals that
manifest in all languages? (3) If yes,
how? Goal My goal is to explore possible answers
to (1)-(3), with neuropsychological justification
offered by Kohno (1993, 2001) and Uemura (1997) .
I propose to define a metrical foot in terms
of a grammatical sense unit, called the PSU, and
a bimoraic foot in terms of a temporal unit,
called a beat, both as substantive universals
there will be simply no exception in that all
languages share in common PSU foot (instead of
stress foot) and bimoraic beat (instead of
bimoraic foot) .
Languages with a principle of end-focus Stress-tim
ed languages (eg. English, German, Swedish,
Russian) Syllables in stress-timed languages are
not evenly timed (S.D.85.7 in Table 1). Thus a
principle of end-focus utilizes a stress to unify
and signal a senseful unit, called a perceptual
sense unit (PSU) (Kohno 1993). Each PSU,
perceptually isochronous in nature, is then
equivalent to a stress foot or stress
group. Syllable-timed languages (e.g.
French, Italian, Hispanic), Syllables in
syllable-timed languages vary less in their
duration (S.D. 74.0 in Table 1), but not
sufficiently uniform to reveal PSUs by
themselves. Thus, a principle of end-focus
using a stress device is as necessary as it is in
stress-timed languages (Dauer 1983). In this
regard, the difference between syllable-timed
languages as stress-times languages is
quantitative in nature and they can be treated as
forming a natural class.
PSU (E1) Thats what John bought
yesterday morning. (E2) Thats the book John
bought yesterday morning. (R1) ja njI xat?u
g?va rjitj ab ?t?m I dont want to
talk about it (examples from Catford 1977, 1988)
(F1) s?-l?-liv?-k?-?ã-a-a?-te-j? C
est le livre que Jean a achete hier
That is the book Jean bought yesterday
(examples
from Koizumi 1996)
Speaker --------gt
Listener neuroarticulatory
neuroperceptual
Languages without a principle of
end-focus Mora-timed languages (e.g.
Japanese) Qualitatively different are mora-timed
languages in which moras are relatively short and
fairly equally timed, and thus mora-counting is
able to reveal the PSU without using stress.
In fact, moraic trochees are largely found as
containing two moras, leading to the notion of
foot-binarity (Hayes 1995). However, it is not
clear how the bimoraic foot S is related to the
PSU.


/ ?


m m
Syllable length (Kohno 2001) Let native speakers
naturally read orally a prose in their language,
record them and measured the length of each
syllable (VOT intervals). A prose contained 243
syllables in Japanese, 107 syllables in English,
128 syllables in Hispanic.
(J1) ko-??-w?-t?-ro-o-?a-kI-no-o-ka-t-t?-ho-n-de
-s? Kore-wa taroo-ga kinoo katta hon
desu This is the book Taro bought
yesterday
Table 1 Result Syllable length average (xms)
and S.D. Japanese X 158,
S.D. 29.5 English X
244, S.D. 85.7 Hispanic
X 201, S.D. 74.0
The notion of beat Kohno (2001) proposed a
fundamental temporal unit, beat, of 330ms, which
controls slot units, pulses such as syllables and
moras. 330ms is a durational boundary to
facilitate holistic sound processing by humans.
All syllabic units of any language occur within
330ms. A beat may contain any number of
moras depending on the speed of speech. In usual
speech it accommodates up to two moras. The
beat, then, is equivalent to a customarily
perceived isochronous unit of the bimorait foot.
In other words, the bimoraic foot is a
neurophysiologically driven grammatical
unit. Correlation between the bimoraic beat and
the PSU The PSU is equivalent to 7 2beats in
length, which in turn conforms to the short term
memory size of 7 2, theorized by Miller (1956).
Neuroperceptual justification for the PSU (Kohno
2001, 2007) The PSU is a unit which
accommodates the amount of materials that can be
processed in our direct short-term memory.
It is a unit driven by the universal human
working memory. Neuroarticulatory justification
for the beat (Uemura 1997) The beat is a
rhythmic unit created by contractions of
respiratory and related laryngeal muscles
to exhale the amount of air to generate one or
more syllables.
Correlations (tentative)
Word Reversing by an Williams patient (Katada
2008) Units reversed are moraic.
Moras and beats are delinked, denoted by / ? .
Bimoraic beats are newly formed. PSU
feet are newly formed.
physiologial motives neuropsychological durational units grammatical units duration
working memory PSU foot 7 2 beats
res. muscle contractions beat moraic trochee 330ms
-- -- syllable/mora varies
articulatory organs -- phonemes varies
Reversed output zagyobate
Foot
ltPSUgt / ?
/ ? /
? B1 B2
B3 ltbeatgt / ?
/ ? / ? m5 m4 m3
m2 m1 (m0) ltmoraicgt /?
/? /? /?
x8 x9 x7 x5 x6 x4 x3 x2 x1 (x0)
ltskeletalgt ? /?
? z a g y o b a t
e ltmelodicgt
Internal structure of PSU foot
tebagyoza ltPSUgt
Foot
/ ?
/ ?
/ ?
ltbeatgt B1
B2 B3
/ ? / ? / ?
ltmoraicgt m1 m2 m3 m4 m5
(m6) /? /?
/? /? ltskeletalgt
x1 x2 x3 x4 x5 x6 x7 x8 x9 (x10)
/? ?
? ltmelodicgt t e b a g
y o z a
Phenomena related to beat ltbased on Kohnos
analysis of 133 English irregular verbsgt The
notion of beat may explain how irregular tense
verbs in English were motivated, namely, English
tried to avoid unstable rhythm of ½ beat.
feeled/sleeped 1.5 beats
go (intr.) ? went goed to/for
1.5 beats felt/slept 1 beat
show(tr.) ?
showed showed NP 2 beats Children move
from child grammar to adult grammar from
neurophysiological necessity.
References (partial) () written in
Japanese Abercrombie , D. (1967) Elements of
general phonetics. Edinburgh Edinburgh
University Press. Catford, J. C. (1977)
Fundamental problems in phonetics. Edinburgh
Edinburgh University Press. Catford, J. C.
(1988) A practical introduction to phonetics.
Oxford Clarendon.  Dauer, R.M.(1983)
Stress-timing and syllable-timing reanalyzed.
Journal of Phonetics 11.51-62 Hayes, B. (1995)
Metrical stress theory principles and case
studies. Chicago University of Chicago
Press. Inaba, S. (1998) Moras, syllables, and
feet in Japanese. Language, Information and
Computation PACLIC12), 12-20 Feb. 1998.
106-117. Kohno, M. (1992) Two mechanisms of
processing sound sentences. In Y. Tohkura, E.
Vatikiotis-Bateson, and Y. Sagisaka (eds.),
Speech Perception, Production and Linguistic
Structure. Oxford IOS Press, 287-296. Kohno, M.
(1993) Perceptual sense unit and choice memory.
International Journal of Psycholinguistics 9-
125. 13-31. Kohno, M. (2001) Recognition
of a spoken language and mechanisms of
generation. Tokyo Kinseido. Koizumi, T. (1996)
An introduction to phonetics. Tokyo Daigaku
Shorin. Liberman, M. (1975) The intonational
system of English. Cambridge, MAMIT
dissertation. Liberman, M. and A. Prince (1977)
On stress and linguistic rhythm. LI
8.249-336. Maekawa, K. (1997) Regions without
accent and intonation accent. In M. Sugito
(1997). 97-124. Miller, G. (1956) The magical
number seven plus or minus two some limits on
our capacity for processing information.
The Psychological Review 63-2.81-97. Poser, W.
J. (1990) Evidence for foot structure in
Japanese. Language 66.78-105. Sugitou, M. (1997)
(chief editor) Accent and intonation of various
dialect. TokyoSanseido. Tateishi, K. (1989)
Theoretical implications of Japanese musicians
language. WCCFL 8. Uemura, Y. (1997) Historical
depth and regional spread of Japanese speech
sounds. In M. Sugitou. 21-62.