Jensen text review

On LinguistList yesterday, Mike Cahill posted a review of John T. Jensen‘s Principles of Generative Phonology textbook, mentioned at least twice here on phonoloblog. The full review is copied below, for your convenience.

Just for the record: I want to encourage (cross-)posting of reviews here on phonoloblog (of stuff that would be of interest to phonologists, obviously). Formal or informal, full book reviews or discussions of recent papers/dissertations/ideas, whathaveyou. This would be a great place to consolidate stuff like that, and for all of us to freely participate in the discussion (for which, of course, you have to register — e-mail me at phonoloblog#gmail|com to find out how).

The following is the text of Mike Cahill’s review found at


If someone is looking for an introductory phonology textbook that lays
out basic rules and phonological reasoning of the SPE-type (Chomsky
& Halle 1968) and does NOT include Optimality Theory, they may
want to consider this book. The author, John T. Jensen (hereafter J),
designed this text for a beginning phonology class. No other linguistics
is assumed, though a phonetics class would be highly useful and an
introduction to morphology also helpful. Each chapter has a set of
exercises at the end.

Chapter 1, on Phonetics, starts out in a straightforward manner. J
justifies the use of segments through such evidence as speech errors,
orthographies, and native speaker intuition. He then presents
consonants in terms of the traditional categories of manner, place,
voicing, airstream mechanisms, and vowels, including specific
discussion on cardinal vowels, in terms of high, mid, low, round, back,
and a bit on diphthongs. For acoustic phonetics, he introduces the
physics of waves and calculation of sound waves leading to formant
frequencies, as well as formant transitions away from consonants, and
displays a spectrogram. Next, he points out several problems with the
IPA, such as the multiplicity of central vowels and the lack of
mnemonicity because of single symbols, rather than diacritics that
would indicate a single place of articulation. For this book, he
proposes his own unique transcription system combining elements of
the IPA, Americanist system, and at least one other.

In Chapter 2, J discusses the central concepts of Contrast and
Distribution, more on distribution than on contrast. He moves quickly
into the notion of complementary distribution, and uses helpful tables
to illustrate this, with the various environments listed in the top row,
and allophones listed along the left. He gives examples from English,
French, and Farsi. He then quite briefly mentions “coincident
distribution” (which others might term “contrast” or “contrastive
distribution”). Anticipating the next chapters, J then talks of the
problem of “overlapping distribution,” where a sound occurs in a
particular environment but could be assigned to either of two
phonemes. He next talks of “pattern congruity”, e.g. that voiceless
stops pattern the same, and mentions the fact that free variation may
also occur. After all this, he introduces the basic rule notation A –> B /
P __ Q, illustrating with r-devoicing in Farsi. J lists and illustrates the
most common types of phonological processes (assimilation,
dissimilation, lenition, fortition, insertion, deletion, lengthening,
compensatory lengthening, shortening). The remainder of the chapter
is taken up with the problems of purely phonemic analysis, especially
neutralization and pattern congruity.

Chapter 3 is on Distinctive Features. J starts with defining them and
showing Turkish back/round vowel harmony in terms of these,
eventually adding more vowel features and more languages like
Lamba. He discusses the feature ATR in both Akan and English as the
same feature (and later on, for Spanish). After this introduction, he
discusses other distinctive features more systematically, basically
repeating the SPE schema. He discusses redundancies in feature
specifications in quite some detail, the goal of such rules being to find
the “minimum set of features” needed to distinguish segments of a
language. For example, he lists ATR in a full feature specification of
Turkish vowels, but redundancy rules account for all ATR
specifications, so he can eliminate that feature from being contrastive
in Turkish.

Chapter 4 covers Alternations and how to write rules to account for
them. J has a very good discussion of Russian Final Devoicing, and
leads the reader agreeably through the logic of choosing one
underlying form over another. His “criteria of phonological analysis”
(predictability, naturalness, simplicity, phonological solutions preferred
to idiosyncratic suppletion) are quite valuable summaries for the
student. The remainder of the chapter expands the details of how to
write rules and the extra notational devices used for these, always
using a real-language example to illustrate. In this, J covers iterative
rules using vowel harmony in Pulaar and Yoruba, uses features
intelligently to combine rules that affect similar sounds in Spanish, and
uses alpha notation for nasal place assimilation in Lumasaaba. He
then takes a time out for a section summarizing five “steps in
phonological analysis”, and a summary of how to write up the analysis
(assuming a class homework assignment rather than a published
paper). These are again valuable sections. His last two sections polish
off the notational conventions of SPE, explaining curly braces,
parentheses, more on Greek variables, angled brackets, mirror image
rules, and transformational rules such as in metathesis, again using
real language examples.

Chapter 5 is on Rule Ordering, which J introduces with a thorough
discussion of Russian Final Devoicing and l-Deletion, introducing the
concept of feeding order with these. Then J adds Dental Stop Deletion
into the mix, showing it bleeds l-Deletion. He discusses iterative rule
application. He then illustrates all these in a quite detailed discussion
of the Spanish trilled and flapped r, and a yet more extensive
discussion of Yawelmani vowel shortening and epenthesis and vowel
harmony. He reviews and systematizes the concepts of feeding and
bleeding, then adding counterfeeding, counterbleeding and opacity,
referring back to previous data on Yawelmani, etc. to illustrate these.

Chapter 6 is on Abstractness, noting at the beginning that any
phonetic transcription is a sort of abstraction, and going on to discuss
in more detail what has been largely assumed before, the notion of
two levels, the underlying being more abstract than the surface. J
notes wryly that the reader “may have thought that the phonological
analyses so far in this book have been anything but simple” but shows
that the criterion of simplicity helps choose one analysis over another.
Another criterion is what J terms the “naturalness condition,” that
underlying representations should be identical to surface ones unless
there is evidence otherwise. Such evidence often comes from
alternations, which J discusses in light of the notion of abstractness.
Generally one of the alternant sounds in question may be considered
the underlying representation (relative concreteness), but J also
discusses the more abstract Yawelmani vowel case, in which an
underlying form /u:/ is posited which never appears on the surface, for
reasons of analytical simplicity and using rules which are needed
independently. He does discuss limits on abstractness, maintaining
that there should be some phonetic relationship between an abstract
underlying form and the surface form. J briefly discusses the
contribution that external evidence (speech errors, second language
acquisition, writing systems, and especially language games and
poetry) may have when positing abstract forms.

Chapter 7 is titled Multilinear Phonology, the first departure from the
SPE framework in the book. The chapter, as might be predicted from
the title, discusses autosegmental and metrical phonology, but is also
a catch-all for other post-SPE developments, including lexical
phonology and underspecification. J starts by showing readers the
matrix of seven-feature system for tone of Wang (1967) applied to the
thirteen Chao tone letters, but then quickly moves into the classical
Mende case of tone melodies which are difficult if not impossible to
express in a strictly linear model such as SPE. He then
autosegmentally illustrates rightward tone shift in Kikuyu, Turkish
vowel harmony, and tone stability.


A very positive point of the book is the abundance of well laid-out exercises at
the end of every chapter. As expected from a text that has been used in class
already, these exercises are quite good for giving practice to the points
covered in the corresponding chapter. One complaint I would have is that several
exercises in Chapter 2 ask the reader to “state a rule,” but only two
examples of rules have been given in the chapter. More examples in
the text would help.

Another point I greatly appreciate is that J is careful not to stray far
from real language data. The book is permeated with good
illustrations of phenomena.

Chapter 4 is perhaps the most valuable in the book. J does quite a
good job in illustrating how to reason phonologically. His discussions
on how to analyze and good argumentation are excellent models to
train students.

However, there is a distinct feeling of datedness to the book. This is a
book that in some ways seems like it was written 20 years ago. SPE is
cited several times as the authority, though J does include what feel
like the new theories of autosegmental, metrical, and lexical

Non-standard terminology in several places will create an
unnecessary barrier to further phonological studies. Examples already
mentioned include J’s idiosyncratic phonetic symbols rather than the
IPA, as well as his use of the term “coincident distribution” rather
than “contrastive distribution.”

The author painstakingly lays out the shortcomings of the phonemic
method, showing the points at which generative phonology provides a
superior analysis, but part of it involves explaining the
morphophonemic level of phonemics. For introductory students, it is
probably not helpful to show all the things that don’t work and that you
are not going to use. J could probably omit this section and content
himself with the summary sentence just before 2.9 that says you can
start with phonemics but it ultimately you need to refer to other
information about the sound patterns of the language.

In introducing autosegmental phonology, there is no discussion of the
formalism and terminology (tier, association line, link etc.). In the first
section J casually throws in the term “toneless suffix” which I would
anticipate would leave the average beginner bewildered. The reader
can probably deduce some of this, but this section stands in contrast
to J’s painstaking definitions and discussions elsewhere.

A few more specific grievances include the fact that some
transcriptions are open to question, as when he transcribes “play” with
both an aspirated [p] and a voiceless [l] in Chapter 2. Also in Chapter
2, J’s 7-line presentation of compensatory lengthening is overly
simplistic in presenting this as a basic process. It would be better to
have omitted this. Also, his first example of shortening is phonetic
rather than phonological: the relative vowel durations in “bead”
and “beat”.

J says in the Preface that he doesn’t want to get into areas where
there is no standard (such as unary features and Feature Geometry),
but this doesn’t stop him discussing distinctive features in Chapter 3,
where there has never been 100% consensus, and the extensive
discussion of redundancy rules is probably overdone as well.

J’s use of ATR (Advanced Tongue Root) as an ersatz vowel height or
tenseness feature for non-African languages is unfortunately not
uncommon, but I wish practitioners of such, in their quest for universal
features, would utilize something different for European languages in
which the tongue root plays no part in contrastive articulation.

Typos are few. In Chapter 2, derivation (37) refers to rule (35) when it
should be rule (36). Something I thought initially was a typo was where
one allomorph of the English plural is listed as [Iz], with the vowel [I] as
in “pick”, rather than the more accepted bar-i. However, the same
vowel is listed as the plural suppletive allomorph in [In] as the plural
in “oxen,” and it would appear that J was trying to simplify for his
students here.

Again, this would be a text that should appeal to some with its basics
on phonological features, rules, and reasoning. Others may be
frustrated with the lack of more recent discoveries and theoretical


Chomsky, Noam, and Morris Halle. 1968. The Sound Pattern of
English. New York: Harper & Row.


Mike Cahill did on-site linguistic investigation in the Konni language of
northern Ghana for several years, including application to literacy and
translation work. He received his Ph.D. from Ohio State University in
1999, and is primarily interested in African phonology, cross-linguistic
patterns in tone, and labial-velar stops and nasals. He currently
serves as SIL’s International Linguistics Coordinator.

2 thoughts on “Jensen text review

  1. Pingback: phonoloblog » Textbook review parallels

  2. Pingback: phonoloblog»Blog Archive » Textbook review parallels

Leave a Reply

Your email address will not be published. Required fields are marked *