Phonology: A Coursebook

Blog post written by Robbert Kennedy, University of California, Santa Barbra

Phonology A Coursebook by Robert Kennedy I am excited to share Phonology: A Coursebook with instructors everywhere. This textbook represents the culmination of many years of thinking about how to make the content of phonology courses more accessible and engaging to students, and I can share a few examples of what is new about it here.

I have always believed phonological analysis to be an important skill for linguists of any stripe, so I think it’s crucial that students establish a solid understanding of its central concepts. But Linguistics is growing as an academic field, with its traditions of structural analysis and documentation joined by those interested in the study of language through the lens of identity, technology, and many other angles. The growth in size and range of our undergraduate population (at my home institution, and surely many others) reflects this. My personal motivation for writing a phonology textbook thus comes from my classroom observation of the varying interests and learning styles among students, not just in phonology courses but in other linguistics courses as well – so that even if the student is not planning on specializing in phonology, they can still experience the course as a practicum in the procedures of the scientific method.

With this in mind, I have structured this book around a mindset of the primacy of data: its chapters are organized around types of phonological processes and patterns, with assimilation, deletion, insertion, harmony, syllabification, stress, and tonal phenomenal all highlighted as objects of phonological analysis. While I have included familiar classic problem sets, including data from languages such as Yokuts, Turkish, Hungarian, Japanese, Kongo, and Polish, I have enriched them with many others that are either less canonical or newly developed, with notable exercises on syllabification, tone, and prosodic morpho-phonology. Moreover, I have used the data to guide the use of formalisms like features, or rules, or tiered representations.

Meanwhile, I have observed in the past that some students have difficulty seeing phonology and the input-output relationship when following the standard teaching practice of introducing them with distributional facts and phonemic analysis. To address this, I introduce the concepts of underlying representations and processes that operate on them with more concretely observable examples of morphophonemic alternation before exploring phonemic analysis and complementary distribution.

 This gives students something more tangible to grasp early on – the idea that a single underlying phoneme could have multiple surface allophones is more plainly obvious when the forms of specific morphemes alternate by their phonological context. In practice, teaching about phonemes by using complementary distribution and mutual exclusivity, which are more circumstantial in their evidentiality, risks a level of abstractness that is perhaps best left until later in the term. There is a parallel to be drawn with calculus, where the instructor may teach either integrals or derivatives first. Teaching derivatives first is more intuitive to many learners, but in phonology it is as if we have been teaching integrals first.

I believe this approach dovetails well with the spirit of Cambridge’s Coursebook series, in which the reader is presented with datasets and exercises, but the analytical steps are narrated procedurally to illustrate the links between detecting patterns and accounting for their nuances and complexity.

The second novel component is a deeper integration of typological generalizations as an element of phonological argumentation. In class when leading students on how to decide among competing analyses, I often find myself turning to typological evidence, yet note that this information is not readily at the hands of undergrads. The organization of the book by processes clarifies that there are certain types of phenomena that are typologically prevalent, and I use this to argue for the student that the formal tools should reflect these trends.

Another deliberate aspect of this textbook is how it treats the role of features and representations. Feature charts and derivational conventions are so rich with detail and precision that students can get lost trying to remember them all, especially if they think of the best analysis as one that uses the correct features.  I often see students struggling to memorize feature charts for IPA symbols rather than thinking of natural classes in more concrete terms. Thus I emphasize in the text that the features are valuable analytical tools, but what a student employs in a given analysis must primarily distinguish groups of sounds that behave differently.

This textbook is aimed at introductory phonology classes, particularly for students who have completed an introductory course in linguistics and/or phonetics and have working knowledge of IPA transcription and some basics of morphological analysis. Nevertheless the datasets are numerous and rich enough to be useful for more advanced students of phonology as well.

I look forward to using this textbook in the classroom and sincerely hope other phonology instructors will find it both useful and engaging as a resource for their students.


Teaching a course on this topic?

EMEA lecturers may request a copy of this title for inspection here

US instructors may request a copy of this title for examination here

Visit the book’s page for more information here

Uptalk: The Phenomenon of Rising Intonation: an interview with author Paul Warren

Paul Warren author of Uptalk 1. Can you define uptalk very briefly for those who don’t know?

Uptalk is the use of rising intonation (voice pitch) at the ends of statements or parts of statements. It is sometimes referred to as the use of question intonation on statements, but this is misleading, because not all questions have rising intonation (indeed there are many question types that tend to have falling intonation, such as those which have a wh-word at the beginning, like who, what, where), and there are rises on statements that are different from uptalk rises (such as on non-final items in a list like apples, oranges, bananas and pears, or the ‘continuation rise’ that you are likely to hear at the comma in Although this has a rise, it is not a question). Typically uptalk, which is also known as upspeak and high rising terminal (amongst other terms), is used to keep an interaction going, inviting the listener into the conversation. This is a specific instance of a more general property of high pitch to show openness, while lower pitch tends to mark finality or closure. However, because rising intonation is frequently associated with questions, many lay observers criticise ‘uptalkers’ for being uncertain about what they are saying. Interestingly, though, studies have shown that uptalk is highly likely in narrative contexts, such as when people are recounting something they have witnessed or experienced firsthand. These are unlikely to be situations where the speaker is uncertain.

   2. What inspired you to write Uptalk?

As a psycholinguist, I devote a lot of my research time to looking at how we produce and understand language, especially spoken language. I have for a long time had a particular interest in how listeners interpret the intonation in utterances that they hear, and when I moved to New Zealand, a country where uptalk has a longer history than in most of the world, I was intrigued by how this particular form of intonation was interpreted. It was clear to me that non-uptalkers frequently arrived at a different interpretation from that intended by the speaker. This interest resulted in a series of research studies, during which I learned more about uptalk in different varieties of English and in other languages too. It seemed a natural next step to put what I had learned into a book where others – whether or not they are linguistics researchers – could have ready access to the wealth of information that is out there concerning the history, spread, and use of uptalk around the world.

   3. How much does it vary according to the speaker’s age, gender and regional dialect?

There are certain parts of the world where uptalk has been a feature of spoken English for quite a long time: New Zealand, Australia and parts of Canada and the United States (particularly California). But it has been reported in many other English-speaking countries, as well as in other languages, particularly either there is where contact with English-speaking communities or a clear influence of the English language on youth culture. Typically, it is associated with young women, but it is by no means exclusively used by females, nor just by the young. Indeed, a number of studies have shown that people of the generation who were young uptalkers in the 1980s have continued to use uptalk as they have grown older. There may be some historical basis for saying that uptalk is a feature of young female speech, since linguists have shown that it is often young women who initiate a change in patterns of language use. Now, however, the claim that young women are the main users of uptalk is probably more a stereotype than a reality. In fact, uptalk is so common in some parts of the English-speaking world that subtle distinctions are developing in what uptalk rises and true question rises sound like, as part of making the difference clearer.

4. What are the key features and benefits that readers will take away from Uptalk ?

What I have tried to do in this book is provide a comprehensive overview of what uptalk is like, including how it differs from other forms of rising intonation; what its many functions and meanings are; how it is distributed and used across the many varieties of English (and other languages) in which it is found; which speaker groups are more likely to use it; and how it is perceived and interpreted by listeners. For those interested in how researchers have investigated uptalk, there is also a chapter on methodology. Because there has been so much discussion of uptalk in newspapers and self-help books, as well as on the radio and television, I also wanted to provide an exploration of the media response to uptalk, including some discussion of the types of statements often used in support of the largely negative claims made by journalists and others. So Uptalk covers a lot of ground, and should be of interest to both linguists and non-linguists alike.

Find out more about Uptalk: The Phenomenon of Rising Intonation


Vowel insertion in Scottish Gaelic

PHO Aug 14Post written by Michael Hammond, Natasha Warner, Andréa Davis, Andrew Carnie, Diana Archangeli and Muriel Fisher,University of Arizona

Based on an article recently published in the journal Phonology

Scottish Gaelic has a process whereby a vowel is inserted into a hetero-organic cluster when the preceding vowel is short, the first consonant is a sonorant, and the second consonant is not a voiceless stop, e.g. arm`army’ /arm/ ->[aram], seanmhair`grandmother’ /ƪɛnvɛr/ -> [ɛnɛvɛr], etc.

These have been cited as instances of excrescent vowels (Hall, 2006). One of the defining properties of such vowels is that they are phonologically inert and are not motivated by-nor do they contribute to-the syllable structure of a language. The basic idea is that excrescent vowels are essentially gestural transitions from one segment to another, without phonological motivation or consequence.

In this paper, we report on a series of six experiments tapping into native speakers’ intuitions of syllable structure in Scottish Gaelic. Our first two experiments tap into whether subjects can distinguish items with inserted vowels from those without. The general result here is that they can.

Experiments #3 and #4 ask subjects to give the number of syllables in a word, either by counting (#3) or by knocking (#4). These show that there is indeed a difference between words with inserted vowels and those without. Interestingly, the results show that inserted vowels are distinct from non-inserted vowels, they are not the same. They contribute significantly to the syllable count, but not as much

as a non-inserted vowel. Finally, experiments #5 and #6 tap into whether the syllabification of intervocalic consonants is affected by the insertion status of the following vowel. Our results show that it is indeed. In general terms, when the following vowel is inserted, the consonant is less likely to affiliate to the left, than to the right.

Our results suggest that the inserted vowels of Scottish Gaelic are not phonologically inert. These vowels contribute significantly and directly to native speaker intuitions, affecting both the number of syllables and the affiliation of consonants to those syllables. Thus, insofar as intuitions about syllable count and the syllabification reflect phonological structure, the inserted vowels of Scottish Gaelic are part of the phonology. However, our results also establish that the relevant vowels have an intermediate phonological status, distinguishing them from underlying vowels as well.

Read the full article ‘Vowel insertion in Scottish Gaelic’ here.