JOHN M .

LIPSKI

On universal phonological features

H u m a n language may be thought of as consisting of a n u m b e r of simultaneous a n d

interrelated levels, all of which contribute t o the process of h u m a n communication.
One of these levels, a n d certainly the most prominent f r o m the standpoint of an
outside observer, is t h a t containing t h e sounds of speech. I n the hands of linguists,
this level becomes known as phonetics, a n d the s t u d y of the audible or phonetic pro-
perties of h u m a n language has long been one cf t h e focal points of linguistic science.
I n particular, the s t u d y of phonetics has brought with it a search for those physiologi-
cal, psychological, a n d acoustic aspects which characterize the communicative possi-
bilities of h u m a n speech. Over t h e years, much scholarly labor has been expended in
describing in great detail the phonetic structures of a large n u m b e r of the world's
languages, a n d almost f r o m the very beginning it became a p p a r e n t to investigators
t h a t certain traits a n d observations occurred again a n d again among the chaos of
sounds represented b y these languages. I t was noticed, for instance, that consonants
a n d vowels t e n d to cluster about certain points of articulation, a n d t h a t some fea-
tures, such as voicing, nasality, aspiration, etc. Avere o f t e n used to distinguish phonetic
segments. Following the discovery of these a n d other recurring properties a m o n g va-
rious languages, it was only a n a t u r a l step to instigate the search for a set of universal
phonetic traits, from the midst of which could be selected particular traits which would
sufficiently describe any phonetic segment which might be encountered. I t is
the p u r p o s e of this p a p e r to explore several theoretical a n d substantive aspects
of t h e search for a universal set of phonetic features, particularly as reflected by
p r o p o n e n t s of generative phonology a n d other modern theories of phonological
investigation."
W h e n a modern phonetician, w h a t e v e r his background, encounters a new segment
which he wishes t o describe, he brings with him a n u m b e r of preconceived notions
a b o u t t h e universality of phonetic traits, based largely on his knowledge of his own
language a n d other languages he has studied. L A D E F O G E D ( 1 9 6 9 ) has noted, t o this
e n d : „At this stage we should p e r h a p s n o t e t h a t we do not gain a n y t h i n g by a t t e m p t -
ing to define features in perceptual terms. I t is impossible t o speak of the percept ion
of phonetic e v e n t s by an unbiased observer, unless we are considering a new born in-
f a n t . A n y b o d y who speaks a language will perceive sounds in t e r m s of his own parti-
cular experience; a n d an i n f a n t ' s innate endowment for perceiving all possible speech
s o u n d s cannot be tested."
T h u s , for example, a Spanish phonetician hearing a sound basically resembling his
own [x], will, immediately classify it as a'voiceless velar fricative', even t h o u g h in
finer detail the new sound m a y be quite different from his own voiceless velar frica-
J . M. LiPSKi. U n i v e r s a l p h o n o l o g i c a l f e a t u r e s 421

tive. 1 The more languages this phonetician has studied which contain dorsal fricatives,
t h e more accurate his description of the new sound is likely to be, but the m a j o r
classificatory features will be basically the same in all instances. It is this f u n d a m e n t a l
similarity between sounds in widely differing languages which has led to the formula-
tion of comprehensive „ s t a n d a r d s " for phonetic description, such as those proposed
by the I n t e r n a t i o n a l Phonetic Association. Such classification s t a n d a r d s make no
claim of universality in the strict sense, but merely provide a set of target points
about which can be clustered a n y sound occurring in the real world. As more langua-
ges are investigated, it is o f t e n found that the existing target points do not adequately
account for certain sounds, and hence new target points are added. From this it may
be seen t h a t it will always be theoretically possible, although methodologically useless,
to arrive at a completely universal set of phonetic traits merely by e n n u m e r a t i n g all
t h e sounds found in all the world's languages. I n fact, this list m a y be e x t e n d e d to its
absurdly logical limit by including every u t t e r a n c e of every speaker, t h r o u g h o u t this
a n d all previous generations. This situation is vaguely reminiscent of N I E T Z S C H E ' S
t h e o r y of eternal recurrance, which s t a t e s in effect t h a t , if one assumes the universe
to be of finite mass, t h e n sooner or later a n y particular combination of this mass
will recur again, in fact, infinitely m a n y times.
A list such as the one mentioned above, although unwieldy and in fact u n a t t a i n a b l e
in practice, would be finite, a n d would assume infinite proportions only if one included
the utterances of all f u t u r e speakers. This, however, would be removing t h e investiga-
tion f r o m t h e domain of science a n d transferring it to t h a t of metaphysics. I t is evi-
d e n t . nonetheless, t h a t such is t h e only method of acheiving universality in its most
restricted sense. I n practice, however, t h e n u m b e r of „universal" categories m u s t be
greatly reduced, in order to secure a system of manageable dimensions. I t is this search
for a meaningful reduction in t h e n u m b e r of phonetic traits which m a y be t a k e n , in t h e
practical sense, as defining the search for a 'universal phonetic system.'
I n practice, the n u m b e r of phonetic categories is immediately reduced b y the gene-
ral limitations on h u m a n aural acuity, which places a constraint on the fineness of
resolution which m a y be obtained between phonetic variants. These limitations, t h e n ,
roughly serve to define a system of „narrow phonetic transcription", whose r u d i m e n t s
have been outlined in such publications as the various I P A handbooks, a n d which has
been f u r t h e r elaborated in more specific studies. Formulating such a narrow phonetic
system basically consists of tabulating all t h e perceptibly different sounds occurring
a m o n g t h e world's languages; a formidable enough task, b u t one which nevertheless
must be completed in order to ensure accurate a n d unambiguous phonetic transcrip-
tions. Such is the position t a k e n , in effect, by C H O M S K Y a n d H A L L E ( 1 9 6 8 : 2 9 4 — 5 )
when t h e y s t a t e : „The total set of features is identical with the set of phonetic pro-
perties t h a n can in principle be controlled in speech : t h e y represent the phonetic
capabilities of m a n . " However, neither in C H O M S K Y a n d H A L L E ' S s t u d y , nor in a n y
other similar work, has this goal been seriously approached, although a n u m b e r of
interesting a n d useful partical system have been proposed. 2
For m a n y aspects of linguistics, even the reduction in size of the set of phonetic va-
r i a n t s a f f o r d e d by a system of narrow phonetic transcription is too cumbersome, with

1
F o r a d i s c u s s i o n o f t h i s p a r t i c u l a r c a s e , s e e HARRIS ( 1 9 6 9 : 191—193).
- A g o o d e x a m p l e is o f f e r e d b y LADEFOGED (1966).
422 J. M. LiPSKi, Universal phonological features

its hundreds of potential categories. Since a large number of linguistic investigators

are interested not in the fine phonetic aspects of speech, but rather in isolating those
aspects which are most relevant for the purpose of communication, it has been felt
that a further reduction in the number of necessary criteria would be desirable. T h e
communicative function of human speech is basically a process of combining members
of a finite set of segments in various w a y s se as to arrive at different meanings; this
is, in effect, the double articulation of MARTINET (1965). Thus, for purposes of studying
the w a y s in which individual meaningful units are built up from sounds, it is merely
necessary to ferret out those particular traits or features inherent in these sounds which
are responsible for signalling difference in meaning; i. e. the distinctive features. It is
the search for these distinctive features which has occupied the time and efforts of
many linguists, particularly since the turn of the century. Surprisingly enough, for a
relatively long period of time, this search was undertaken on a language-by-language
basis, with little explicit attention being given to the possibilities of finding a universal
set of distinctive features. The main reason for this apparent lack of interest with
universality seems to be the tacit assumption, on the part of most linguists, that the
features they were working with were universal, since they were applied again and again
i n a variety of languages. Once attention was specifically directed toward the actual uni-
versality of distinctive features, however, it became increasingly clear that the
existent theories were inadequately formulated, being basically a collection of
language-specific observations, and hence it was felt by some to be necessary to get at
the root of the problem concerning exactly what features serve to distinguish human
utterances.
Again, the problem of isolating the universally available traits seems on the surface
to be a fairly simple one: one simple combines the results of a phonological analysis
of all known languages and extracts from this total accumulation each feature which
serves, in one or more languages, to effect a difference in meaning. Preciesly this task
was undertaken by many investigators, in an imperfect but nonetheless useful w a y ;
the most noteworthy example being provided by TRUBETZKOY ( 1 9 3 9 ) . Under such early
„universal" systems, one classified vowels in terms of such features as front, back,
round, nasal, etc., while consonants were categorized b y features like stop, fricative,
labial, velar, and so forth. Once this was done, one could go on to investigate the man-
ner in which these „distinctive features" were combined to form that minimal distinc-
tive element known as the phoneme. TRUBETZKOY found, for example, that most pho-
nemic oppositions could be divided into major categories, according to the extent to
which other groups of phonemes in the same language shared the same distinctive
features. These ideas were further elaborated by MARTINET ( 1 9 5 5 ) in his theory of
„phonological economy", which stated that oppositions which recurred at several points
in a language ; i. e. which formed a „correlation", would become aligned in sucha fashion
as to make maximal use of the distinctive traits among themselves. Some of MARTINET'S
epigones carried these notions to even further extremes: for instance DORFMAN (1968)
proposed that the major correlations align themselves under ideal conditions so as to
form a square. While there are many difficulties inherent in a MARTINET-like wiew of
phonological structure, such a system does have the noted merit of formulating and
making explicit certain important aspects of phonology. Perhaps the single most
important fact noted by adherents of structuralist phonology is the existence
of at least a partial hierarchy among the distinctive traits or features involv-
J . M. LIPSKI, Universal phonological features 423

ed. :i Thus, for example, vowels a n d consonants were first of all s e p a r a t e d and t r e a t e d
as two f u n d a m e n t a l classes, although o f t e n related in various ways. Within these
classes, f u r t h e r hierarchization generally took place, although generally along lan-
guage-specific lines. For instance, one language may contain voiced and voiceless
stops, but only voiceless fricatives (i. e. Spanish); in these cases one would assume
t h a t a correlation of closure was higher in the hierarchy of distinctive features t h a n
t h a t of voicing. Similarly, in a language which distinguished five different points of
articulation for o b s t r u e n t s , but only three for nasals, it would be deduced t h a t
nasality was lower down on the distinctive hierarchy t h a n place of articulation. Such
s t r u c t u r a l i n t e r p r e t a t i o n s do not, however, lend themselves easily to questions of
universality, since b y their very n a t u r e they are concerned with phonemic interactions
within a given language. Whereas some of the features or correlations involved seem
to be u n d i s p u t a b l y of a universal n a t u r e , such as voicing, nasality, vocalicity, etc., in
finer points it is not immediately a p p a r e n t t h a t it is possible, even in theory, t o
arrive at universal notions within the structuralist theory, except in the trivial sense
listed above. For linguists of structuralist persuasion, this is not necessarily a serious
d r a w b a c k , since describing t h e t r a i t s of individual languages is a legitimate a n d reward-
ing pursuit in its own right ; f u r t h e r m o r e , it o f t e n appears, in examining various struc-
t u r a l descriptions, t h a t there i s a certain limit beyond which universality is impossible.
W h e t h e r there is indeed a limit on the a m o u n t of universality which can be e x t r a c t -
ed from phonological descriptions of languages is at the present time u n k n o w n , al-
though it seems likely t h a t given the interest exhibited by scholars in various discipli-
nes t h e answer will eventually be uncovered. I t should be noted, however, t h a t t h e
possibility of finding a non-trivial a n d meaningful set of features which m a y be consi-
dered effectively universal must be regarded as a scientific hypothesis, t o be s u b j e c t e d
to intensive search, a n d to be evaluated in the light of each new piece of information
t h a t comes f o r t h .
I t is p e r h a p s an u n f o r t u n a t e fact t h a t many modern linguists have accepted as
axiomatic the premise t h a t there exists a small class of a t most a couple dozen distinct-
ive features which may be considered for all intents a n d purposes universal. T h u s ,
for example, C H O M S K Y a n d H A L L E ( 1 9 6 8 ) speak repeatedly of „universal phonetics",
although at no point is a n y comprehensive a t t e m p t m a d e to j u s t i f y the use of such a
term, either theoretically, or substantively. Their feelings, however, are typical of a
great n u m b e r of c o n t e m p o r a r y linguists, who feel t h a t a 'universal'phonetic (or phono-
logical) system as outlined above is much too cumbersome a n d „uneconomical" for use
in describing h u m a n language a n d who have opted instead, for no other explicitly
a p p a r e n t motive t h a n descriptive simplicity, for a small set of (as yet undiscovered)
f e a t u r e s or t r a i t s which represent t r u e phonological universale.
A key point in this connection is exactly what intrinsic phonetic content the distinct-
ive f e a t u r e s are presumed to have. For m a n y investigators the distinctive features
have been regarded as totally a b s t r a c t a n d a r b i t r a r y classificatory devices,
which merely served t o signal oppositions between phonemes, b u t which did n o t
necessarily have a meaningful interpretation in the real world. T h u s B L O O M F I E L D
(1933: 129—30) r e m a r k e d : „A list or table of t h e phonemes of a language should there-
fore ignore all non-distinctive features . . . . such lists or tables are usually made on t h e

3
See also the discussion in LADEFOGED ( 1 9 6 8 : 2 8 5 ) .
424 J . M. LIPSKI, U n i v e r s a l p h o n o l o g i c a l features

basis of practical-phonetic classifications . . . tables like these, even when t h e y exclude

non-distinctive features, are nevertheless irrelevant to the structure of the language
because t h e y group the phonemes according to the linguist's notion of their physio-
logic character, a n d not according to the p a r t s which the several p h o n e m e s play in
t h e workings of the language."
As a n additional example, NIDA (1949: 31) found it quite unobjectionable t o classi-
f y t h e f o u r - m e m b e r MAZATEC vowel system a s :
front back
high
low
More recently, such a position has been t a k e n by FUDGE (1967) who n o t e s (p. 7): „If
t h e r e are no other bases for establishing such [i. e. phonemic] relationships, acoustic
similarity will be considered, t h o u g h a n y other t y p e of p a t t e r n i n g which is indicated
will be t a k e n as more i m p o r t a n t . "
I t is immediately evident, however, t h a t if one accepts the notion t h a t distinctive
f e a t u r e s have no intrinsic phonetic c o n t e n t ; i. e. t h a t t h e y have no measurable
physical correlates, t h e r e is virtually no way t o constrain the theory so as t o eliminate a
seemingly infinite n u m b e r of bizarre a n d arbitrarily constructed „correlations". F o r
e x a m p l e , given a n y r a n d o m collection on η phonemes, one could define a n u m b e r of
suitable f e a t u r e s which would formally distinguish t h e m , without giving the slightest
clue as t o their actual relationships in the language in question. Even a r g u m e n t s based
on phonemic p a t t e r n i n g leave an excessive degree of freedom, for, especially a m o n g
t h e consonants, it is possible to discover entire groups of segments which are only slight-
ly phonetically related, b u t which exhibit an identical behavior with respect t o p a t t e r n -
ing. I t is only b y requiring the additional criterion of phonetic similarity t h a t a b s u r d
a n d meaningless descriptions m a y be ruled out ; a n d precisely this f a c t is noted b y
FUDGE, as cited above. Once phonetic criteria are brought t o bear, however, all argu-
m e n t s as to t h e totally a b s t r a c t s t a t u s of distinctive features vanish, a n d it becomes
necessary to assign physical correlates t o the f e a t u r e s being employed. F o r this reason,
it seems, in fact, t h a t t h e complete abstractness of distinctive features is somewhat of
a n overidealization, a n d consequently t h a t whatever system of f e a t u r e s is employed
will h a v e to reflect observable d a t a in t h e real world.
This view has been sustained, at least superficially, by most p r o p o n e n t s of the theo-
r y of generative phonology, which assumes as one of its basic t e n e t s the existence of a
small set of universal distinctive features t h a t m a y be used in formulating phonologi-
cal rules for a n y language. HALLE (1954:203—4) s t a t e s : „In conformity with our require-
m e n t t h a t our t e r m s be empirically meaningful, t h e „questions" [i. e. distinctive fea-
tures] of the preceding p a r a g r a p h are of a kind t o which answers can be provided by
physical m e a s u r e m e n t ; i. e. t h e y are questions regarding the presence of certain
definite physical properties."
S i m i l a r l y , t h e s t u d y o f JAKOBSON, F A N T a n d H A L L E (1951) p r o v i d e s a n e l a b o r a t e , a l -
t h o u g h in t h e long r u n inconclusive, a t t e m p t at defining t h e physical correlates of
one proposed set of features. On t h e other h a n d , a n overabundance of phonetic detail
can lead t o a n overcomplicated system in which t r u e phonological processes become
h i d d e n in the midst of r e d u n d a n t phonetic detail. CHOMSKY a n d HALLE (1968: 296)
n o t e , t o this e f f e c t :
J. M. LiPSKi, Universal phonological features 425

„ W e t h e r e f o r e can represent lexical i t e m s n e i t h e r in phonetic t r a n s c r i p t i o n nor in a n

a r b i t r a r y n o t a t i o n totally u n r e l a t e d to t h e e l e m e n t s of t h e p h o n e t i c t r a n s c r i p t i o n .
W h a t is n e e d e d is a r e p r e s e n t a t i o n which falls between these t w o e x t r e m e s . "
F i n d i n g such a n i n t e r m e d i a t e r e p r e s e n t a t i o n is clearly easier said t h a n done, how-
ever, as o n e m a y easily observe by looking over the various p r o p o s a l s concerning fea-
t u r e s y s t e m s which h a v e a p p e a r e d in t h e last few years. I n f a c t , as n o t e d a b o v e , t h e
basic premise t h a t t h e r e exists such a level of r e p r e s e n t a t i o n which m a y be discovered
in a n empirical fashion is itself in n e e d of justification. Such a j u s t i f i c a t i o n has gener-
ally not b e e n f o r t h c o m i n g ; b u t instead a curious sort of d u a l i t y t o w a r d t h i s q u e s t i o n
h a s b e e n m a n i f e s t e d b y m a n y linguists. F o r e x a m p l e . H A L L E (1962) s p e n d s t h e g r e a t e r
p o r t i o n of his p a p e r a t t e m p t i n g to j u s t i f y his choice of distinctive f e a t u r e s a n d t h e n
concludes his s t u d y by saying in e f f e c t t h a t all t h i s doesn't m a t t e r a n y w a y , since a t t h e
d i s t i n c t i v e level t h e f e a t u r e s h a v e o n l y a classificatory s t a t u s . T h i s s t a t e m e n t is a p p a r -
e n t l y in conflict b o t h w i t h H A L L E ' S earlier r e m a r k s quoted a b o v e , a n d w i t h t h e pre-
viously c i t e d s t a t e m e n t f r o m t h e Sound Pattern of English, a n d it d e m o n s t r a t e s t h e
g e n e r a l level of insecurity concerning t h e existence of a smallset of universal phonologi-
cal f e a t u r e s .
A n o t h e r q u e s t i o n which m u s t be a d d r e s s e d , given t h e a p p a r e n t l y necessary a s s u m p -
t i o n t h a t d i s t i n c t i v e f e a t u r e s m u s t h a v e a certain measure of intrinsic p h o n e t i c c o n t e n t ,
is w h e t h e r t h i s p h o n e t i c c o n t e n t be specified in acoustic or a r t i c u l a t o r y t e r m s , or p e r -
h a p s as a c o m b i n a t i o n of b o t h . I n early schemes, exemplified b y t h o s e of T R U B E T Z K O Y ,
M A R T I N E T a n d P I K E , p r i m a r y a t t e n t i o n w a s p a i d to a r t i c u l a t o r y p a r a m e t e r s , t h e s e
being most easily accessible, a l t h o u g h it was implicitly a d m i t t e d t h a t t h e f e a t u r e s also
h a d acoustic correlates, since most p h o n e t i c investigation was d o e n b y listening t o spo-
k e n s p e e c h . As e a r l y a s 1 9 3 8 , however, J A K O B S O N ( 1 9 3 9 ) was a d v o c a t i n g a phonologi-
cal s y s t e m b a s e d on p e r c e p t u a l (i. e. acoustic) p a r a m e t e r s . C e r t a i n f e a t u r e s , such a s
voicing, n a s a l i t y , friction, etc. seem t o serve equally well a s e i t h e r a r t i c u l a t o r y o r
acoustic f e a t u r e s . W i t h r e g a r d t o f i n e r distinctions, however, t h e r e h a s so far b e e n n o
m e t h o d of a c c u r a t e l y specifying a one-to-one correlation b e t w e e n a r t i c u l a t o r y a n d
acoustic f e a t u r e s t o describe t h e p r o d u c t i o n of consonants a n d vowels.
D u e largely t o J A K O B S O N ' S pioneering work, a new era of p h o n e t i c s w a s u s h e r e d in,
where a c o u s t i c f e a t u r e s were a s s u m e d t o p r e d o m i n a t e a t t h e phonological level.
P e r h a p s t h e b e s t - k n o w n a t t e m p t at d e f i n i n g a set of acoustically-based distinctive
f e a t u r e s is o f f e r e d b y J A K O B S O N , F A N T a n d H A L L E ( 1 9 5 1 ) . I n t h i s w o r k a n e f f o r t is
m a d e t o e s t a b l i s h for each p r o p o s e d f e a t u r e a set of physical correlates, b a s e d e i t h e r
u p o n t h e a c o u s t i c p r o p e r t i e s of speech as displayed on a s p e c t r o g r a p h , or u p o n t h e
physical p r o p e r t i e s of the vocal t r a c t , considered as a r e s o n a n t c a v i t y . Such an investi-
gation is r e n d e r e d e x t r e m e l y difficult by t h e f a c t t h a t , acoustically speaking, s p e e c h
is a c o n t i n u u m of s o u n d which m a y not be readilly segmented i n t o discrete u n i t s s u c h
as p h o n e m e s . Such, in fact, is n o t e d b y H A L L E ( 1 9 5 4 : 1 9 7 ) , w h o t h e n goes on t o a d d :
„ A l t h o u g h t h e view of language as a c o n t i n u o u s p h e n o m e n o n is simple a n d s t r a i g h t -
f o r w a r d f r o m a s t r i c t l y physical s t a n d p o i n t , it has certain i n h e r e n t difficulties w h i c h
m a k e it u n d e s i r a b l e as a basis for descriptions, a n d investigators of language . . . h a v e
usually p r e f e r r e d t o describe language as a sequence of discrete e v e n t s . F u r t h e r m o r e
it is not necessary t h a t a physical p h e n o m e n o n be actually d i s c o n t i n u o u s in order t o
break it u p into a sequence of discrete e v e n t s . I t is possible t o divide it i n t o s e g m e n t s
if we can s h o w e x a c t l y how it is t o be d o n e . "
426 J . M. LiPSKi, Universal phonological features

I n retrospect, -it seems as t h o u g h the acoustic f e a t u r e s proposed b y J A K O B S O N ,

FANT a n d H A L L E did not do a n a d e q u a t e j o b of u n a m b i g u o u s l y s e g m e n t i n g the chain
of speech, as m a n y subsequent investigators were quick t o point o u t . F o r example,
t h e f e a t u r e flat was supposed to s t a n d for a whole g r o u p of s e c o n d a r y articulations,
including r o u n d i n g (for vowels), labialization (for consonants), retroflexion, velariza-
tion, pharyngealization, etc. These a d d e d t r a i t s show few if a n y similarities, either
acoustically or articulatorily, other t h a n t h e r a t h e r v a g u e fact of being „articulated
with a secondary constriction in the periphery of the vocal t r a c t " . F u r t h e r m o r e , diffi-
culties arose in t r y i n g t o a p p l y other localization f e a t u r e s , such a s grave, diffuse, a n d
compact t o b o t h consonants a n d vowels. F o r vowels, such f e a t u r e s m a y be specified
(although not necessarily with concomitant p e r c e p t u a l justification) in t e r m s of rela-
tive f o r m a n t frequencies. Most t r u e consonants, however, especially stops, have no
inherent f o r m a n t s t r u c t u r e , a n d one has t o deal instead with f o r m a n t shifts indicating
the transition between a consonant a n d a vowel. These shifts, however, generally do not
g r o u p consonants with vowels in t h e m a n n e r p r e d i c t e d b y t h e J A K O B S O N - F A N T - H A L L E
f e a t u r e s ; f u r t h e r m o r e , t h e y are highly d e p e n d e n t on t h e contiguous vowels. 4 T h u s ,
in order t o j u s t i f y t he classification of vowels a n d c o n s o n a n t s by m e a n s of a single set
of features, one must make reference t o t h e various physical configurations assumed
b y t h e vocal t r a c t (as is done, for example, b y H A L L E 1 9 6 2 ) , which in effect presents
a r e t u r n t o a r t i c u l a t o r y criteria.
Observing t h e obvious inadequacies i n h e r e n t in t h e system of acoustic features as
proposed in Preliminaries, m a n y linguists began a d v o c a t i n g a r e t u r n t o articulation-
based distinctive p a r a m e t e r s . L A D E F O G E D ( 1 9 6 6 : 2 9 5 ) n o t e s t h a t t h e use of a set of
articulatory f e a t u r e s allows an intrinsic p o r t r a y a l of „impossible" feature com-
binations, while in a wholly acoustic f r a m e w o r k t h e s e restrictions must be
s t a t e d explicitly, in addition t o the basic definitions of t h e f e a t u r e s : „this is
because all traditional phonetic charts are a r r a n g e d so as t o preclude a n u m b e r
of possible combinations of states of t h e glottis a n d m a n n e r of articulation".
K I M ( 1 9 6 6 ) p r o p o s e d a n articulatory f r a m e w o r k , vaguely reminiscent of the
I P A - t y p e a r t i c u l a t o r y phonetics, in which to p o r t r a y all possible phonological
combinations. These same concepts were f u r t h e r r e f i n e d in various details in KIM
( 1 9 6 8 , 1 9 7 0 ) , a n d h a v e been neatly summarized, along with some p e r t i n e n t physiologi-
cal d a t a , in K I M ( 1 9 7 1 ) . T h e e x t e n t t o which t h e r é é v a l u a t i o n of phonological featu-
res in t e r m s of a r t i c u l a t o r y correlates h a s t a k e n hold a m o n g c o n t e m p o r a r y linguists
m a y be seen b y t h e fact t h a t , in t h e most recent m a n i f e s t o of phonological theory,
C H O M S K Y a n d H A L L E ( 1 9 6 8 : Chap. 7 ) h a v e r e t u r n e d t o a f u n d a m e n t a l l y articulatory
f r a m e w o r k . T h e y s t a t e (p. 303) t h a t : „the complete identification of vowel a n d conso-
n a n t features seems in retrospect t o have been too radical a solution." I n the discus-
sion which follows this s t a t e m e n t , C H O M S K Y a n d H A L L E proceed t o outline some of the
difficulties e n c o u n t e r e d b y the J a k o b s o n i a n acoustic f e a t u r e s , most of which h a d alrea-
d y b e e n p o i n t e d o u t b y o t h e r writers. 5
Looking over t h e linguistic scene of the p a s t t h i r t y years or so, t h e r e seems t o have
been a sort of Zipfian oscillation between a r t i c u l a t o r y a n d acoustic f e a t u r e systems.
4
See, for example F I S C H E R - J O R G E N S E N ( 1 9 5 4 ) , L I B E R M A N et al ( 1 9 5 6 ) , MALMBERG
(1955). These works also include references to other pertinent studies.
5
A comprehensive critique of some of the early distinctive features is offered by Mc-
CAWLEY ( 1 9 6 7 ) .
J . M. LiPSKi. Universal phonological features 427

E a c h n e w a t t e m p t at s u p p l y i n g a d i f f e r e n t f e a t u r e s y s t e m was realized by m e a n s of a
sort of technological o n e - u p m a n s h i p ; a n d given t h e a d v a n c i n g s t a t e of t h e a r t , t h e r e
is no reason t o d o u b t t h a t t h e c u r r e n t CHOMSKY-HALLE a r t i c u l a t o r y f e a t u r e s will o n e
d a y be o v e r t h r o w n in a coup d'état by some enterprising linguists ( p e r h a p s e v e n
CHOMSKY a n d H A L L E themselves), a r m e d with a n impressive a r r a y of n e w a c o u s t i c
d a t a . I n f a c t , e v e n utilizing t h e p r e s e n t l y available d a t a , one m a y discover a n u m b e r
of difficulties in e i t h e r f r a m e w o r k . T h i s fact has a p p a r e n t l y led H A L L ( 1 9 6 9 : 2 1 1 ) t o
d o u b t t h e overall v a l i d i t y of d i s t i n c t i v e f e a t u r e s in general. S u c h a position, h o w e v e r ,
r e p r e s e n t s a bit of an o v e r r e a c t i o n , since it a p p e a r s obvious to m o s t i n v e s t i g a t o r s t h a t
t h e concept of d i s t i n c t i v e f e a t u r e s m a y be quite useful in m a n y cases. I n a n e x t r e m e l y
p e r c e p t i v e article, D E L A T T R E ( 1 9 6 7 ) h a s given a c o m p r e h e n s i v e r e v i e w of investiga-
t i o n s p e r t a i n i n g t o b o t h t h e a r t i c u l a t o r y a n d the acoustic c o r r e l a t e s of d i s t i n c t i v e
f e a t u r e s , w i t h a n u m b e r of i n t e r e s t i n g conclusions. I t is shown, f o r e x a m p l e , t h a t cer-
t a i n p h o n e t i c p h e n o m e n a , such a s v e l a r a r t i c u l a i ion in stops, exhibit a r t i c u l a t o r y correla-
t e s which a r e simpler a n d m o r e n e a r l y ' i n v a r i a n t , ' while o t h e r p h e n o m e n a , including
d i s t i n c t i v e vowel n a s a l i z a t i o n a n d t h e various t y p e s o f r-sounds i n A m e r i c a n E n g l i s h ,
e x h i b i t f a i r l y c o n s t a n t a c o u s t i c characteristics, b u t considerable a r t i c u l a t o r y v a r i a -
bility. F r o m this, D E L A T T R E c o n c l u d e s t h a t n e i t h e r acoustical n o r a r t i c u l a t o r y d a t a
s h o u l d b e u s e d in d e f i n i n g t h e d i s t i n c t i v e f e a t u r e s , b u t r a t h e r t h a t t h e f e a t u r e s
s h o u l d b e c o n s i d e r e d o n l y a t t h e „ p e r c e p t u a l " level. These conclusions a r e p r e s e n t e d
in a passage w o r t h q u o t i n g a t l e n g t h (pp. 2 3 - 4 ) : „If it c a n n o t be s h o w n t h a t distinc-
tive f e a t u r e s are closer t o o n e s o r t of o b j e c t i v e correlate t h a n t o t h e o t h e r , it is
p e r h a p s t h a t i n v a r i a n c e e x i s t s n e i t h e r a t t h e acoustic nor a t t h e a r t i c u l a t o r y level.
T h e n if t h e r e a r e s u c h signals a s i n v a r i a n t s b y m e a n s of which d i s t i n c t i v e f e a t u r e s
f u n c t i o n in t h e process of s p e e c h recognition, these i n v a r i a n t s m u s t f u n c t i o n o n l y
a t t h e p e r c e p t u a l level a n d m u s t m e d i a t e b e t w e e n ft distinctive f e a t u r e a n d i t s o b j e c -
t i v e correlates. W h e t h e r t h e i n v a r i a n t s are n a m e d in acoustical, in a r t i c u l a t o r y or in
p e r c e p t u a l t e r m i n o l o g y is of little i m p o r t a n c e . T h e y would p e r h a p s f a r e best a s
n u m b e r s , a n d t h e d i s t i n c t i v e f e a t u r e s a s letters of t h e a l p h a b e t . B u t t h e r e s h o u l d be
n o o b j e c t i o n t o occasionally using c o n s e c r a t e d t e r m s such as n a s a l i t y or voicing for
t h e d i s t i n c t i v e f e a t u r e s , a s long a s t h e y are n o t misleading, a n d t o u s i n g e i t h e r a n a r -
t i c u l a t o r y t e r m or a n acoustic one, whichever is simpler, for t h e i n v a r i a n t s . "
D E L A T T R E ' S r e m a r k s a r e r a t h e r d i f f i c u l t to e v a l u a t e , since t h e t e r m „ p e r c e p t u a l " a s
he uses it a p p e a r s t o b e u n d i s c o v e r a b l e . If distinctive f e a t u r e s c a n n o t b e d i s c o v e r e d
b y a p p l y i n g e i t h e r a c o u s t i c or a r t i c u l a t o r y criteria, it is h a r d t o i m a g i n e e x a c t l y w h a t
does d e t e r m i n e t h e intrinsic c o n t e n t of t h e f e a t u r e s . Using a d i a g r a m t o i l l u s t r a t e his
t h e o r y , D E L A T T R E s u g g e s t s (p. 2 4 ) t h a t „ t h e objective acoustic a n d a r t i c u l a t o r y corre-
l a t e s a r e r e l a t e d t o a d i s t i n c t i v e f e a t u r e only indirectly, b y m e a n s of s u b j e c t i v e i n v a -
r i a n t s . " I n e f f e c t , t h i s s t a t e m e n t n e c e s s i t a t e s t h a t we s t a t e , for e x a m p l e , t h a t a s p e a -
k e r recognizes t h e p h o n e m e / d / s i m p l y because he perceives it as / d / , which places
us r i g h t b a c k in t h e p r e - d i s t i n c t i v e f e a t u r e days. P a r t of the c o n c e p t u a l i m p a s s e e n g e n d -
e r e d b y D E L A T T R E ' S p r o p o s a l s s t e m s f r o m t h e failure to consider t h e relative a s p e c t s
of s p e e c h p e r c e p t i o n . I t h a s b e e n s u f f i c i e n t l y d e m o n s t r a t e d , b y D E L A T T R E a n d o t h e r s ,
t h a t it is well-nigh impossible t o e s t a b l i s h an absolute set of correlates f o r m o s t distinc-
t i v e f e a t u r e s , using either a c o u s t i c or physiological d a t a . A large a r e a of speech p e r c e p -
tion, h o w e v e r , is c o n c e r n e d w i t h isolating a given sound relative t o o t h e r s o u n d s in t h e
s p o k e n c h a i n . T h u s , for e x a m p l e , L A D E F O G E D ( 1 9 6 7 ) has s h o w n t h a t t h e p e r c e p t i o n
28 Z. Phonetik 5 / 7 4
428 J . M. LiPSKi, Universal phonological features

of vowel phonemes is dependent upon having heard enough of the discourse to be

able to establish the relative positions of the phonemes with respect to one another.
Similarly, it has been sho\vn that the formant transitions associated with stop conso-
nants are largely dependent on the surrounding vowels, with certain characteristics
being obliterated in the environment of certain vowels. 6
It is quite possible that, in the long run, D E L A T T B E ' S pessimism about the feasibility
of finding an exclusively articulatory or exclusively acoustic set of correlates for dis-
tinctive features will turn out to be justified. However, before completely abandoning
the investigation, it is useful to consider the possibility of two simultaneous coexistant
sets of features, one primarily articulatory and one predominantly acoustic. This would
not be particularly surprising in view oft he fact that the processof human communication
consists of two fundamental phases: that of speaking and that of listening. Therefore,
it might be surmised that a speaker would use one set of features, based on articula-
tory parameters, when speaking, and another set of features, this time acoustically
defined, when listening to the speech if others. This possibility seems to be hinted at in
the first few pages of F U D G E (1967), where the articulatory and the auditory components
are separated for discussion. It turns out, however, that F U D G E seeks to acheive syn-
thesis between the two modes of speech in a different fashion ; namely by making all
the features totally abstract: „The logical conclusion of this is that phonologists
(above all, generative phonologists) ought to burn their phonetic boats and turn
to a genuinely abstract framework. By so doing they will escape the fate of not
only falling between two stools . . . but also ending up sitting in the very place which
they have expended such strenuous and well-justified efforts to avoid." (p. 26)
F U D G E ' S conclusions notwithstanding, the possibility of two simultaneous sets of
features should be pursued further. First of all, let us assume, intuitively enough, that
when a speaker produces an utterance he utilizes a set of articulatory features. Due to
the inherent auditory feedback present in all normal individuals, he will be able to
hear his own speech as he is producing it. This auditory feedback mechanism is presum-
ably part of the same component used in listening to the speech of others, and hence
presumably relies on predominantly acoustic features. Between the auditory and the
articulatory features, then, there must exist a series of transfer functions which map
between the acoustic and the articulatory components. These mappings are not gener-
ally one-to-one, nor are they necessarily onto mappings. For example, the speaker
may be able to produce the same acoustic result by means of several different articu-
latory gestures. Similarly, he may encounter a sound for which he has no articulatory
analog; i. e. which does not exist in his own speech. In the act of speaking, then, the
articulatory features will serve to form the utterance, while the acoustic features inhe-
rent in the auditory feedback mechanism will serve to „monitor" the produced
speech.
In the act of listening, the above situation may be considered as reversed. The
received sounds are processed in terms of their acoustic features, but for most indivi-
duals there seems to exist a (sometimes quite imperfect) feedback to the articulatory
features. Thus, in listening to a person speaking with a lisp, many people testify to
actually being able to 'feel' the anomaly in their own vocal apparatus. Once again, the
conversion of acoustic to articulatory features is not a perfect one; thus, upon hearing

6 Cf. DELATTRE (1967) and the references given therein.

J . M. LiPSKi, U n i v e r s a l phonological features 429

a given sound, there is often doubt in the listener's mind as to how it was produced.
The model of speech production and recognition proposed above may be schematically
represented roughly as in Figure 1.

Spoken speech
acoustic articulatory
features C features

!
1 1
1 1
ι ι
11 I1
1 I !!
1 1
To phonological To phonological
__component transfer transfer component
functions functions
1 1 1
ii1 1 1
1
1
1
1
1
11 1 1
1 1
Spoken speech
articulatory acoustic
features features

Speaker A Speaker Β

As it stands, the above model is purely hypothetical, since little investigation has
gone into the possibility of using both acoustic and articulatory parameters to describe
the same situation. I n view of the fundamental duality inherent in human language,
however, it appears certain t h a t some such situation will have to be envisaged in order
to account for the problems encountered in using strictly physiological or strictly
acoustic features. One very promising area for future research along these lines could
be the domain of language change. Considering, for the moment, only those sound chan-
ges which are apparently spontaneous; i. e. which do not result from morphological or
analogical contamination, borrowing, substrata, etc., one might hope to discover sound
changes whose primary motive was acoustic in nature, while other changes would
appear to have been triggered by physiological factors. A prime target for investiga-
tion would be changes involving consonant clusters. For example, early Vulgar Latin
had both the medial cluster -tl- (as in vet(u)lus) and the medial cluster -cl- (as in
oc(u)lum). Now the acoustic differences bet ween -tl- and -cl- are quite small, whereas
from an articulatory standpoint no particular difficulty seems to be involved in either
case. I n late Vulgar Latin, however, the cluster -tl- became -cl-·, thus vetlus > veclus.
I t may be due to the same factors t h a t the modern Romance languages do not contain
the cluster *tl in native words, although pi, cl, f l , etc. freely occur.
The reduction of certain Latin consonant clusters in Italian seems to point to a
pressure in terms of articulatory parameters. To cite a single example, the Latin
clusters -pt- and -ct- merged in Italian as the geminate -tt- ; hence scriptum > scritto,
odo > otto. etc. Acoustically, the clusters -pt-, -et-, and -tt- are quite distinct, and may
be easily perceived by speakers of modern Italian. From a physiological standpoint,
however, a considerable economy in articulatory effort was acheived by changing
clusters with two differing points of articulation into a single geminate with a con-
stant point of articulation.
Many scholars, inclduing A N D E R S O N ( 1 9 6 5 ) , P U L G R A M ( 1 9 7 0 ) , K L A U S E N B U R G E R
(1972), etc. have expended great efforts to demonstrate that the emergence of new
28*
430 J . M. LIPSKI, Universal phonological features

c o n s o n a n t clusters in a language is largely d e t e r m i n e d b y t h e s t r u c t u r e of the a l r e a d y

e x i s t i n g clusters. More s p e c i f i c a l l y , t h e y h a v e a t t e m p t e d to d e m o n s t r a t e t h a t poten-
tial n e w c l u s t e r s ; i. e. clusters t h a t do not o c c u r but w h i c h might o c c u r due to s u c h
f a c t o r s as s y n c o p e , t e n d to be 'dissolvable' in t e r m s of the s t r u c t u r e of the g i v e n
l a n g u a g e . T h a t is, t h e cluster must be able to be split into an initial consonant (or
c o n s o n a n t s ) w h i c h o c c u r s w o r d - f i n a l l y plus a s e c o n d c o n s o n a n t ( s ) w h i c h occurs word-
initially. T h e m o t i v a t i o n for these restrictions s e e m s b y a n d large t o be a r t i c u l a t o r y in
n a t u r e , since m a n y p o t e n t i a l but non-dissolvable c l u s t e r s are a c o u s t i c a l l y quite distinct
a n d m a y b e p e r c e i v e d b y s p e a k e r s o f the l a n g u a g e in q u e s t i o n w i t h o u t d i f f i c u l t y . F o r
e x a m p l e , S p a n i s h has, a m o n g others, the following p o t e n t i a l nondissolvable clusters:
*vl, *pf, *gb. *kf. e t c . Spanish speakers can distinguish t h e s e clusters without pro-
blem, a n d most c a n e v e n u t t e r t h e m , in meaningless n o n s e n s e s y l l a b l e s . T h e s e clusters,
h o w e v e r , d o not f o r m part o f s p o k e n Spanish, nor is t h e r e a n y indication t h a t t h e y
will d o so in t h e forseeable f u t u r e . T h e reasons for this seem t o be a r t i c u l a t o r y , as
n o t e d a b o v e . T h i s is not t o s a y t h a t t h e r e is a n y i n h e r e n t p h y s i o l o g i c a l d i f f i c u l t y in
u t t e r i n g such s e q u e n c e s ; r a t h e r it seems that d u e t o t h e set o f n e u r o m u s c u l a r h a b i t s
w h i c h t h e S p a n i s h s p e a k e r has acquired, these c l u s t e r s s e e m s t r a n g e a n d foreign t o
h i m . O f course, t h e role of a c o u s t i c f e a t u r e s in this r e s p e c t c a n n o t be o v e r l o o k e d , since
a c o u s t i c a l l y t h e s e clusters are classified as s t r a n g e b y S p a n i s h speakers. T h e f a c t
r e m a i n s , ho\vever, t h a t is g e n e r a l l y easier for a s p e a k e r of a n y l a n g u a g e to perceive
s e q u e n c e s of s o u n d s w h i c h are foreign to him t h a n it is for him t o a c t u a l l y titter these
sequences.
I t m a y well be, t h e n , t h a t t h e s e p a r a t i o n of p h y s i o l o g i c a l and a c o u s t i c features m a y
be v e r i f i e d b y o b s e r v i n g l a n g u a g e change. S u c h an e n t e r p r i s e is n o t an e a s y one, for
d e a l i n g w i t h h i s t o r i c a l linguistics brings f o r t h a host o f i n d e t e r m i n a c i e s w h i c h o f t e n
c a u s e e v e n t h e b e s t a n a l y s i s to f o u n d e r . P r e s u m i n g , h o w e v e r , t h a t t h e p a r t i a l separa-
tion o f the t w o s e t s o f correlates can be d e t e r m i n e d , b y w h a t e v e r means, there still
r e m a i n s the p r o b l e m of d e t e r m i n i n g the n a t u r e o f t h e f u n c t i o n s w h i c h m a p b e t w e e n
t h e a r t i c u l a t o r y a n d t h e a c o u s t i c f e a t u r e s . A t this s t a g e o f t h e i n v e s t i g a t i o n , such f u n c -
t i o n s are m e r e l y a sort o f u n d e t e r m i n e d t r a n s f e r m e c h a n i s m , m u c h like the v a r i o u s
„ b l a c k b o x e s " w h i c h i n h a b i t e d m u c h of CHOMSKY'S e a r l y w o r k . One fact t h a t does
seem o b v i o u s , a t least a t t h e p r e s e n t time, is t h a t t h e s e f u n c t i o n s m u s t be language-
s p e c i f i c . T h i s is so since, a l t h o u g h there m a y v e r y well be a u n i v e r s a l set of f e a t u r e s ,
e a c h l a n g u a g e c h o o s e s f r o m a m o n g these f e a t u r e s , a n d c o m b i n e s t h e m in a u n i q u e
f a s h i o n . T h i s is t o s a y t h a t e a c h language has its o w n set o f o p p o s i t i o n s , b o t h articula-
t o r y a n d a c o u s t i c , a n d likewise e a c h language i n v o l v e s a set o f f e a t u r e s w h i c h m a y b e
considered p r o p e r l y r e d u n d a n t . T o f u r t h e r c o m p l i c a t e t h e p r o b l e m , each i n d i v i d u a l
s p e a k e r s e e m s t o e x h i b i t a s l i g h t l y d i f f e r e n t b e h a v i o r w i t h regard to distinctive
f e a t u r e s , w h i c h suggests t h a t investigation o f t h e c o n v e r s i o n f u n c t i o n s b e t w e e n
p h y s i o l o g i c a l a n d acoustic p a r a m e t e r s might best be c o v e r e d b y t h e domain of indivi-
d u a l p s y c h o l o g y . I n a n y e v e n t , t h e possibility o f using t w o s i m u l t a n e o u s sets of f e a t u -
res s h o u l d be l e f t o p e n , since therein lies the p r o m i s e o f r e s o l v i n g m a n y o f the p r o b l e m s
w h i c h d i s t i n c t i v e f e a t u r e t h e o r y has e n c o u n t e r e d to d a t e .
J . M. LIPSKI. Universal phonological features 431

References

ANDERSON. J a m e s M. 1965. A study of syncope in Vulgar Latin. Word 21, 70—85.

BLOOMFIELD. Leonard. 1933. Language. New York: Holt.
CHOMSKY, Noam A. and Morris HALLE. 1968. The Sound Pattern of English. New Y o r k :
Harper & Row.
DELATTRE. Pierre. 1967. Acoustic or articulatory invariants'! Glossa 1, 3—25.
DORFMAN. Eugene. 1968. Correlation and core-relation in diachronic Romance phonology.
W o r d 24. 8 1 - 9 8 .
FISCHER-JORGENSEN, Eli. 1954. Acoustic analysis of stop consonants. Miscellenea Phone-
t i c a 2. 4 2 - 5 9 .
FUDGE. E . C. 1967. The nature of phonological primes. Journal of Linguistics 3, 1—36.
HALL. Robert Α., J r . 1970. Some recent developments in American linguistics. Neuphilologi-
sche Mitteilungen 70, 192-227.
HALLE. Morris. 1954. The strategy of phonemics. Word 10. 197—209.
—. 1962. On the bases of phonology, in : The Structure of Language, ed. J . A. FODOR and J . J .
KATZ, pp. 3 2 4 - 3 3 3 .
HARRIS, J a m e s W. 1969. Spanish Phonology. Cambridge: Μ. I. T. Press.
JAKOBSON, Roman. 1939. Observations sur le classement phonologique des consonnes. P a p e r
read before t h e Cercle Linguistique de Prague, 1938; later presented a t the 3rd
International Congress of Phonetic Sciences (Ghent, 1939).
JAKOBSON, R o m a n , C. GUNNAR M . FANT, a n d M o r r i s HALLE. 1951. Preliminaries to Speech
Analysis. Cambridge: M. I . T. Press.
KIM, Chin-Wu. 1966. The linguistic specification of speech. Unpublished P h . D. dissertation,
U. C. L. A.
—. 1968. When liquids become glides. Chicago Journal of Linguistics 2, 45—50.
—. 1970. Two phonological notes: A # a n d B b . Unpublished paper distributed by I n d i a n a
University Linguistics Club.
—. 1971. Experimental phonetics. I n : A Survey of Linguistic Science, ed. W . O. Dingwall.
U. of Maryland Press, pp. 16—135.
KLAUSENBÜRGER, J ü r g e n . 1972. French Prosodies and Phonotactics. T ü b i n g e n ; Max Nie-
meyer.
LADEFOGED, Peter. 1966. The nature of general phonetic theories. Reprinted in the George-
town University R o u n d Table Selected Papers in Linguistics 1961 — 1965, pp. 283—298.
Georgetown University Press, 1968.
—. 1967. The nature of vowel quality. I n : Three Areas of Experimental Phonetics, p p . 50—
142. London: Oxford University Press.
—. 1969. The phonetic framework of generative phonology. Unpublished MS: U. C. L. A.
LIBERMAN, Alvin M., Pierre DELATTRE, L. J . GERSTMAN, F. S. Cooper. 1956. Tempo of
frequency change as a cue for distinguishing classes of speech sounds. J o u r n a l of Experi-
mental Psychology 52, 127-137.
MALMBERG, Bertil. 1955. The phonetic basis for syllable division. Studia Linguistica 9, 80—
87.
MARTINET, André. 1955. Economie des changements phonétiques. Berne: Francke.
—. 1965. La linguistique synchronique. P a r i s : Presses Universitaires de France.
MCCAWLEY. 1967. Le rôle d'un système de traits phonologiques dans une théorie du langage.'
Langages 8, 112-23.
NIDA, Eugene A. 1949. Morphology. Ann Arbor: U. of Michigan.
PULGRAM, Ernst. 1970. Syllable, Word, Nexus, Cursus. The H a g u e : Mouton.
TRUBETZKOY, Nicolai. 1939. Grundzüge der Phonologie.