Daniel Silverman (1997), Phasing and Recoverability. In the series
Outstanding Dissertations in Linguistics. Garland Publishing, New
York. Pp. xiv, 242.
Reviewed by Kimary Shahin, Birzeit University/University of British
This book, Silverman's 1995 UCLA dissertation, examines the role of
articulatory timing ('phasing') in the maintenance of contrasts in sound
systems. The author's main claim is that articulatory gestures are
sequenced in specific ways so cues to the segment's distinctiveness are
recoverable by the listener. He argues that languages use optimal
phasing relationships except where prohibited and suboptimal
relationships only if they use optimal ones. Data from several
languages are discussed. Supporting spectrograms are
This review briefly summarizes the book, then critiques it from a
narrow and a broad perspective. From a narrow perspective, I find
Silverman's phasing observations enlightening and his arguments with
respect to auditory representations convincing. From a broad
perspective, the book is not entirely clear because the author's
assumptions with respect to the relation between phonetics and
phonology are not spelled out. However, it is through less
constrained approaches such as Silverman's that new insights into
phonetics vs. phonology will likely emerge. Discussion of this
book is timely, given 1999's four phonetics-phonology meetings: HIL
(Leiden, January), GLOW (Potsdam, April), ChiPhon (Chicago,
April), ICPhS (San Fransisco, July).
2. Summary of the book
Chapter 1, "Introduction" (3-25), introduces the phasing relationships
gestural sequencing, expansion, and truncation and illustrates how
each can effect auditory salience, hence recoverability. Chapter 2,
"Previous work" (27-49), summarizes and criticizes previous approaches
to phasing. Silverman uses a version of Browman & Goldstein's (e.g.,
1986) Articulatory Phonology. Chapter 3, "Obstruents and laryngeal
gestures" (51-82), explains the phasing relationships that optimally
cue aspiration and glottalization for obstruents. Silverman argues
that in Chong the morphology forces suboptimal phasing for
glottalization. Chapter 4, "Sonorants and laryngeal gestures"
(83-107), explains the phasing relationships that optimally cue place
of articulation and glottalization for nasals and breathy and creaky
voice for glides. Liquids are argued to have relatively free phasing.
Chapter 5, "Vowels and laryngeal gestures" (109-211), discusses how
breathy voice, creaky voice, and tone are optimally recoverable for
vowels. Contrastive phonation is claimed to be implemented serially
with modal phonation for vowels that also contrast for tone, based on
data from Otomanguean languages. Chapter 6, "Concluding remarks"
(213), reiterates the main claim of the book. It is followed by a
list of references and an index.
3.1. Data analysis and generalizations
This book presents enlightening explanations of the auditory, and
articulatory, aerodynamic, and acoustic bases of several timing
relationships between laryngeal, supralaryngeal, and even respiratory
gestures. For example, in the Hebrew hitpa?el pattern, a tsV sequence
is metathesized to stV (p.11). Silverman argues that this is
motivated by the more optimal release of the stop into a vowel;
fricatives, by contrast, are less dependent on such a release. This
suggests to me a phasing basis for the unique license of English /s/
as first C in a CCC onset cluster. Also fascinating is the analysis
of Comaltepec Chinantec 'ballistic accent' (p.157ff). The accent was
previously analyzed as a complex of pitch, amplitude, duration, and
aspiration properties but Silverman reduces it to an aspiration
contrast. Further interesting discussions concern why non-high front
vowels are more likely to spread rounding than high front vowels, why
preaspirated stops are rare, and why postaspirated stops and breathy
vowels don't contrastively co-occur. (See p.14, 55, and 55/64, for
the respective explanations.) Silverman's arguments with respect to
auditory representations, which are based on straightforward claims
about auditory nerve response to various acoustic inputs, are general
in nature and convincing. (Johnson 1997:49-62 on auditory phonetics
is a good background read here.) His diagrams of phasing
relationships and auditory response are effective.
This book has two potential weaknesses. First, the arguments
regarding phasing and recoverability are apparently not always based
on observations from instrumental data. This gives the book a
sometimes speculative feel (e.g., p.142: "For breathiness, vocal fold
tension should be decreased... Similarly, glottal aperture might be
reduced for pitch increases... breathiness may be accompanied by
larynx lowering"). More instrumentation is needed to confirm some of
the claims and generalizations. Second, Silverman frequently equates
breathy voice with aspiration with h, and creaky voice with glottal
stop. He does not outline his assumptions with respect to the
phonological status - segment, or secondary articulation - of glottal
frication or constriction in various contexts, nor does he discuss
whether, and why or why not, that status is important. As
phonologists usually crucially distinguish segment vs. secondary
articulation, it seems there's an important issue left unaddressed
Finally, 'perceptual salience' is described (p.4) as "not necessarily
quantifiable". However, it seems that if it is defined in terms of
rate auditory nerve firing as the author suggests (p.5), it should be
straightforwardly quantifiable. Exciting instrumentation no doubt
3.2. The broader picture
There is a growing literature on the role of auditory salience in
phonology. Silverman's book is a very important foundational work in
this new field (see also, e.g., Steriade 1995, 1997 and Hume
One fairly standard approach says that where there is an auditory
basis for a phonological contrast, that basis is the phonetic
grounding (Archangeli & Pulleyblank 1994), in this case auditory, for
the phonological patterning; for more grounding examples, see, e.g.,
Archangeli & Pulleyblank (1994), Hume (forthcoming), Jiang-King
(1996), and Shahin (1997). The discrete cognitive units of phonology
are distinguished from phonetic tokens, which are the physical
implementations of phonological constructs. Phonological analysis is
distinguished from phonetics excursus on the physical support for
phonological claims. See Shahin (1997) for extensive discussion of
these important issues, including criteria for determining the
phonetic or phonological status for a given sound property, based on
several previous works (e.g., Flemming 1995, Henke 1966, Kiparsky
1985, Liberman 1983, Liberman & Pierrehumbert 1982, Mohanan 1982,
hman 1966, Pulleyblank 1986, Steriade 1995, 1997).
Silverman describes his book (p.xi) as a phonological study. However,
it is largely concerned with physical details: the movement of
articulatory structures, and aerodynamic, acoustic, and auditory
properties. In fact, he states (p.46): "I have not found patterns
that require a segmental analysis." By the phonetics vs. phonology
distinction described in the preceding paragraph, the book would be
classified as a phonetic study, albeit one concerned with the physical
properties of speech sounds from a cognitive perspective, with the
goal of determining how linguistic information is organized (Rischel
1991). However, it is hard to know for sure, because Silverman does
not clarify his assumptions with respect to the nature of phonological
units, and the phonetics vs. phonology distinction. Consider Flemming
(1995), which provides extensive evidence for the acoustic bases of
several sound patternings. Flemming claims that the phonology
contains, i.e., the mind stores, acoustic representations. Does
Silverman assume we store phasing details? The big question here is:
Is there still a general economy assumption in phonological theory?
Do we assume phonetic enrichment only where there is no tenable
segmental analysis (Steriade 1995, 1997)? (See Flemming 1995 for a
convincing case involving retroflection.)
Despite Silverman's unclarity, it is clear that studies such as
Silverman's explore relationships that might be ignored under a more
standard approach. Because of this, it is through such studies that
our understanding of phonology and the role of phonetics in phonology
will likely advance, towards the answers to the big epistemological
questions that remain (see Rischel 1991): Are the phonology and
phonetics separate units? If so, where is their interface - before
the phrasal (postlexical) phonology or after? Or are they one unit?
If so, do they overlap, is there a continuum, or are they
non-distinct? What is the nature of the cognitive constructs
Archangeli, D. & D. Pulleyblank (1994), Grounded Phonology.
Cambridge, MA: MIT Press.
Browman, C. P. & L. Goldstein (1986), "Towards an articulatory
phonology. Phonology Yearbook 3:219-252.
Flemming, E. (1995), Auditory Representations in Phonology. Ph.d.
Henke, W. (1966), Dynamic Articulatory Model of Speech Production
Using Computer Simulation. Ph.d. dissertation, MIT.
Hume, E. (forthcoming), "The Role of Perceptibility in
Consonant/Consonant Metathesis". In S. J. Blake, E.-S. Kim, & K.
Shahin, eds., Proceedings of the West Coast Conference on Formal
Linguistics 17. Stanford: CSLI.
Jiang-King, P. (1996). Tone-Vowel Interaction in Optimality Theory.
Ph.d. dissertation, UBC. (ROA-266-0698 re-entitled An Optimality
Account of Tone-Vowel Interaction in Northern Min)
Johnson, K. (1997), Acoustic & Auditory Phonetics. Cambridge, MA:
Kiparsky, P. (1985), "Some consequences of lexical phonology".
Phonology Yearbook 2:85-138
Liberman, M. (1983), "Phonetic representations". Paper presented at
the Stanford Workshop on Lexical Phonology and Morphology.
Liberman, M. & J. Pierrehumbert (1982), "Intonational invariance
under changes in pitch range and length". Bell Labs ms.
Mohanan, K. P. (1982), Lexical Phonology. Indiana University
hman, S. (1966), "Coarticulation in VCV utterances: spectrographic
measurements". JASA 39:151-168.
Pulleyblank, D. (1986), Tone in Lexical Phonology. Dordrecht: D.
Reidel Publishing Co.
Rischel, J. (1991), "The relevance of phonetics for phonology: a
commentary". Phonetica 48:233-262.
Shahin, K. (1997), Postvelar Harmony: An Examination of its Bases
and Crosslinguistic Variation. Ph.d. dissertation, UBC.
Steriade, D. (1995), "Laryngeal neutralisation and laryngeal features".
Talk at the U Arizona Conference on Featural Relations.
_____ (1997), "Phonetics in phonology". UBC talk.
Kimary Shahin is Assistant Professor at Birzeit University Dept. of
English and Researcher with UBC Linguistics. Her interests include
phonology, phonetics, and phonological acquisition.
Dept. of English
P.O. Box 14, Birzeit
West Bank, via Israel
Dept. of Linguistics
University of British Columbia
Buch. E270, 1866 Main Mall
Canada V6T 1Z1
Get Your Private, Free Email at http://www.hotmail.com