Page 1
University of Babylon
College of Education for Human Sciences
Department of English
FFuunnccttiioonnaall PPhhoonnoollooggyy AA TThheeoorreettiiccaallllyy DDeessccrriippttiivvee AAccccoouunntt wwiitthh aann
““EEppiilloogguuee”” oonn FFoorrmmaall PPhhoonnoollooggyy
A Presentation for
A course in Phonetics and Phonology
PhD Programme, 2012-2013
First Semester
By
Ahmed Sahib Jabir
Under the Supervision of
Prof. Fareed H. Al-Hindawi, PhD
November, 27th , 2012
Page 2
Contents
Section Page
1. Functional phonology
1. 1. Phonological Models
1. 2. Functional Principles
1. 2. 1. Functional principles of speech production
1. 2. 2. The functional principle of the communication channel
1. 2. 3. Functional principles of speech perception
1. 3. Functions of phonic elements
1.The representative function
2. The indexical function
3. The appellative
4. The culminative function
5. The distinctive function
6. The contrastive function
7. The delimitative function
8. The expressive function
1. 4. The Functional Model of Phonology
1. 5. The Functional load
2. Formal phonology
3. Conclusions
References
Page 3
1. Functional phonology
“Functional phonology” is the type of phonology, part of
functional linguistics, which is generally associated with the
Linguistic Circle of Prague; hence it is sometimes referred to as
Prague School Phonology. In the literature, the term functional is
usually used to contrast with the term formal. In this respect, Davis
and Baertsch (2012: 8) affirm that a formal explanation of phonology
denotes that a process occurs in a particular language because of the
way the constraints are ranked or ordered. A functional explanation,
however, references something outside the language system. This
could include allusion to general cognitive abilities such as
perception or to frequency among other factors.
Although it is difficult to specify an exact date for the
beginnings of the theory of functional phonology, it is claimed that its
early beginning was in 1911 when Daniel Jones went to St Petersburg
and there he met Ščerba, a professor of French and follower of
Baudouin de Courtenay. From that meeting on, there were attempts
to distinguish the phoneme from a speech sound (Akamatsu, 1992:
iv). Nevertheless, its “real” start is usually associated with the Prague
Linguistic Circle which was founded in 1926. At that time, Jakobson
and Trubetzkoy attached great importance to the oppositions among
phonemes rather than to the phonemes themselves. Thus to say that
English has phonemes /s/ and /z/ is a statement about a distinction
which English speakers make and recognize rather than a claim
about phonemes as mental images or phonetic entities.
(http://www.ling.fju.edu.tw/phono/prague.htm)
Page 4
As hardly anything is static, functional phonology has
undergone a substantial development, either during the work of the
members of the LCP themselves or by adoption and ameliorations of
its concepts by the Functionalist School led by Martinet (Bičan 2005:
6). In the recent years, two names are much in the center of the
theory, Tsutomu Akamatsu whose book Principles of Functional
Phonology (1992) is introduced by Martinet; and Paul Boersma the
whose publications (his PhD dissertation (1998) as well as a number
of papers and studies1997-2000) all elaborate on the concepts and
principles of functional phonology.
1. 1. Phonological Models
Until fairly recently, phonology was dealt with only within two
competing models. The first is the structuralist model which has a
discrete phonological level that is transitional between two
sequentially ordered modules associated with phonology and
phonetics:
(1) Structuralist model of phonology
| underlying | → / phonemic / → [ phonetic ]
The second is the generative model collapsing the phonological and
phonetic modules into one, which means it rejects the discrete
phonological surface level.
(2) Generative model of phonology
| underlying | → [ phonetic ]
Boersma (1999: 1), however, argues that “these two models actually
share the major assumption of hybrid cognitive phonological
Page 5
representations” and thus proposes a third model called functional,
claiming that it draws closer on the reality of the phonological
phenomena. That model has an articulatory phonetic level lying
between two sequentially ordered modules associated with
production and perception:
(3) Functional model of phonology
| underlying | → [ phonetic ] →/ phonemic /
He (ibid) explains that the functional model of phonology in (3)
is expressed in the structuralist and generative terms for maximal
comparability with the structuralist and generative models.
Consequently,(4) is a paraphrase of (3) by using functional terms that
make explicit the distinction between articulation and perception:
(4) Functional model of the production grammar
|perceptual specification| → [articulatory implementation] →
/perceptual output/
To illustrate the difference between the generative model and
the functional model of phonology, Boersma (1999: 6) cites an
example of /s/ production before and after teeth loss:
Figure (1) Generative view of teeth loss
Page 6
Figure (2) Functional view of teeth loss
1. 2. Functional Principles
Language is intended to convey information from one person to
another as quickly and clearly as possible. As such, each of its aspects
will have to do with the three elements of production, channel, and
perception (Passy, 1890 cited in Boersma, 1997: 3). Therefore, the
functional principles will in accordance with these three elements.
1. 2. 1. Functional principles of speech production
As far as the production of speech is concerned, two principles are
significant, these are: (1) the principle of economy stating that:
“languages tend to get rid of anything that is superfluous” and (2) the
principle of emphasis which proclaims that: “languages tend to
stress or exaggerate anything that is necessary” (Passy, 1890 cited in
Boersma, 1997: 3).
The use of the terms superfluous and necessary in these two
principles emphasizes the idea that articulatorily stimulated
Page 7
constraints should be adhered to if they do not violate the (more
important) perceptually stimulated constraints. Clearly these
principles harmonize with the speaker-oriented principle of the
minimization of articulatory effort and the listener-oriented principle
of the maximization of perceptual contrast.
1. 2. 2. Functional principle of the communication channel
In so far as the communication channel is concerned, there comes out
the principle of the maximization of information flow which says:
“put as many bits of information in every second of speech as you
can”, provided that such bits of information are communicative and
not redundant.
1. 2. 3. Functional principles of speech perception
Regarding speech production, two functional principles emerge:
maximization of recognition and minimization of categorization.
Through the first principle, the listener attempts to maximally utilize
the accessible acoustic information, because that will help them
identify the meaning of the utterance. Alternatively, by the second
principle the disambiguation of an utterance is facilitated by having a
lesser number of perceptual classes into which the acoustic input can
be scrutinized.
1. 3. Functions of phonic elements
Akamatsu (1992: 17) mentions that functional phonology is
typically concerned with identifying, describing, and classifying
various functions accomplished by phonic elements in a certain
language. He (ibid: 17-24) states the following main functions:
Page 8
1. The representative function is the function that relates the
linguistic sign to the referent (hence it is also called the referential
function). Speakers inform listeners of whatever extralinguistic
facts or states they are talking about. For example, intonation
contours provide an interpretation for a sentence by indicating
which part of the information is viewed as new/known, salient
/less salient or topic/comment…etc.
2. The indexical function is the one which helps identify the
speaker as belonging to different social groups (middle class,
working class, …etc.), geographical places (urban, rural …etc. ),
and occupations (lawyer, sergeant, …etc.) (Malmkjar 2002: 160).
3. The appellative function is that which serves to provoke well-
definable impressions or feelings in the listener as when uttering
an imperative with an intonation inducing the listener not to
comply (ibid) as when insincerely inviting someone for dinner!
4. The culminative function is the function which helps indicate
how many significant units (i.e. “words” or “word-combinations”)
there are in an utterance. According to Akamatsu (1992: 20) this
function is manifested in Spanish, for instance, where polysyllabic
words are stressed on only one of the syllables so the number of
stressed syllables will identify the number of polysyllabic words
in an utterance.
5. The delimitative function is the function which spots the
boundary between significant units. Akamatsu (1992: 21) states
that this function is realized in languages like Czech, Finnish, and
Estonia where the first syllable of polysyllabic words is normally
stressed, the matter which demarcates words (hence this function
is alternatively called demarcative).
Page 9
6. The distinctive function is a function which derives directly
from the concept of phonological opposition. It is the function by
virtue of which linguistic forms are opposed to, or differentiated
form, each other. The minimal linguistic form that is meaningful,
or the minimal significant unit, is known as a moneme, which
consists in the association between a signifier (vocal expression)
and a signified (semantic content). For example, in English ‘bit’
and ‘bet’ are monemes semantically distinguished through the
phonetic difference.
7. The contrastive function enables the listener to analyze a
spoken chain into a series of significant units like monemes,
words, phrases, etc. Stress in any language functions contrastively
by bringing into prominence one, and only one, syllable in what is
called an accentual unit. What is meant by the term contrastive is
that the stressed syllable contrasts with the unstressed syllable
and characterizes the accentual unit as a whole. In this respect,
Akamatsu (1992: 23) states that in languages like
8. The expressive function is that function whereby speakers
convey to listeners their state of mind without resorting to the
use of additional monemes. Saying “That tree is eNNNNormous!”
instead of saying “That tree is really/absolutely/extremely
enormous!” is an example of this function. (162)
1. 4. The Functional Model of Phonology
After being differentiated from the structural and generative
models, the functional model will now be elaborated on to make
more explicit the processes of speech production and
comprehension. These processes, Boersma (2000: 1) states, are best
Page 10
described with reference to three “Optimality-Theoretic grammars”
shown in Figure (3) below.
Fig. (3) The grammar model of functional phonology (after Boersma 2000)
First there is the production grammar which starts with an
underlying form that is shaped by means of certain perceptual
specifications to form a continuous articulatory output which is then
transformed by the speaker’s perception system to a more distinct
perceptual output form. From a list of appropriate output candidates,
the chosen one will be that which minimally violates the ranked
constraints of the production grammar.
The constraints are divided into two groups: articulatory
constraints (ART), which evaluate each articulatory output, thus
implementing the functional principle of minimizing articulatory
effort, and faithfulness constraints (FAITH), which evaluate the
similarity between each perceptual output and the underlying form,
thus implementing the functional principle of minimizing perceptual
confusion.
Page 11
Secondly, the perception grammar consists of constraints that
help to classify the acoustic input to the ear into a finite number of
perceptual categories (anti-categorization constraints CATEG,
perceptual faithfulness constraints WARP) and higher-level structures
(obligatory contour principle OCP, and line-crossing constraints LCC).
This grammar was mentioned above as a part of the speech
production process, but is also used by the listener as a first step in
the comprehension process.
Thirdly, the recognition grammar maps the discrete output of
the perception grammar to underlying lexical forms. It consists of
constraints that evaluate the lexical and semantic appropriateness of
recognized underlying forms (*LEX) and, as in the production
grammar, faithfulness constraints (FAITH).
1. 5. The Functional load
The term functional load refers to the case in which certain
features make contrasts in a language and how much use that
language makes from such contrasts. Also referred to as phonemic
load, in phonology it is used to denote the measure of the work which
two phonemes do in keeping utterances apart. Put differently, it is
the gauge of the frequency with which two phonemes contrast in all
possible environments (Surendran and Niyogi, 2003:1).
The first suggested measurement for functional load was the
number of minimal pairs, but this does not take into account word
frequency and is difficult to generalize beyond binary phonemic
oppositions. Hockett (1955) proposed an information theoretic
Page 12
definition which has since been generalized. Now, given a large text
corpus, one can compute the functional load of any phonological
contrast including distinctive features, suprasegmentals, and
distinctions between groups of phonemes (Wikipedia).
In English for example, vowels have a very high functional load.
There are numerous sets of words which are distinguished just by
their vowels, such as:
pin, pen, pun, prn, pein, pgin, pn:n
but, bet, bit, bi:t, beqt, bgit, beit, bn:t
Voicing is similar, as can be seen in pad - bad, sue - zoo.
Speakers who do not control these differences make it very difficult
for others to understand them. Although voicing in English is
important, its difference between the two fricatives written ⟨th⟩, /θ,
ð/, has a very low functional load: it is difficult to find meaningful
distinctions dependent solely on this difference. One of the few
examples is thigh vs. thy although the two can be distinguished from
context alone, not to mention that thy is no longer in normal use in
English. Similar is the difference of /dʒ/ (written ⟨j⟩, ⟨ge⟩, etc.)
versus /ʒ/ (resulting from /z + j/, or the ⟨j⟩, ⟨ge⟩, etc. in some
recent French loanwords), as in virgin vs. version.
In this regard, Surendran and Levow (2004) have found that
the functional load of tone in Mandarin, a Chinese dialect, is as high
as that of vowels. This means the tone (which is a property of
syllables in Mandarin) is as important to identify as it is to identify its
vowels.
Page 13
2. Formal phonology
Formal phonology is that view of phonology which refers to the
conception of taking phonology to be a formal object in the sense of
having abstract formal properties. The most eminent work in this
regard is Chomsky’s who denies that human language is designed for
communication and holds the view that language is designed for
thinking (this is why he is considered as anti-functionalist).
The term “formal phonology” is also used to denote any set of
formal devices for the representation of linguistic structure, such as
the formalisms known as metrical grids used in metrical phonology.
Some linguists take the term to denote a way of describing human
languages in terms of mathematical or logical formalisms. Therefore,
some have argued that Chomsky’s work is not properly formalist, in
the sense that it is not properly mathematical in nature.
In the relevant literature, “formal phonology” is the term used
to encompass any theory that does not adopt the functionalist view
(i. e.) it refers to theories that are concerned rather with “form” than
with “function”. Consequently, atomic phonology, autosegmental
phonology, computational phonology, generative phonology, linear
phonology, stratificational phonology…etc. are all considered part of
formal phonology.
Page 14
3. Conclusions
Functional phonology is the view of studying phonology from
the angle of the functions of phonic elements of a language and how
they are employed to express different “meanings” and reflect
various attitudes and feelings. Thus, it focuses on function rather than
form.
It is usually associated with the linguistic circle of Prague
because its pioneers are the founders of this circle; namely,
Trubetzkoy and Jakobson who were the first to evolve this view.
Other figures who developed it through its course of progress are
Martinet and, more recently, Akamatsu and Boersma.
A essential difference between functional phonology and other
theories is its distinction between “phonetic implementation” and
“perceptual representation”. It believes that an underlying form is
phonetically implemented under the “supervision” of the perceptual
representation.
Page 15
References
Akamatsu, T. (1992) Essentials of Functional Phonology: with a
forward by Andre Martinet. Peeters Louvain-la-neuve:
Leuven.
Bičan, A. (2005) http://www.phil.muni.cz/linguistica/art/bican/bic-
001.pdf
Boersma P. (1997) “Elements of Functional Phonology”
--------------- (1999) “Nasal harmony in functional phonology.”
http://www.fon.hum.uva.nl/paul/
--------------- (2000) “Phonetically-driven acquisition of phonology”
---------------(1998). Functional phonology: formalizing the
interactions between articulatory and perceptual drives.
PhD dissertation, University of Amsterdam. LOT
International Series 11. The Hague: Holland Academic
Graphics. [http://www.fon.hum.uva.nl/paul/diss/]
Davis S. and Baertsch K. (2012) “Formal versus Functional
Explanation for a Universal Theory of Syllable Structure:
The Case of Vowel Epenthesis in Winnebago.” Journal of
Universal Language 13-2 (PP 7-34).
Hockett, C. (1955) A manual of phonology International journal of
American linguistics, Vol. 21, No. 4, Part 1 Pp. v, 246
http://en.wikipedia.org/wiki/Functional_load
Page 16
Malmkjar, K. (2002) The Linguistics Encyclopedia, Second Edition
http://books.google.iq/books?id=XMaq84Mc_5sC&pg=PA1
60&lpg=PA160&dq=%22the+appellative+function+of+pho
nic+elements%22&source=bl&ots=7iEPkds1wo&sig=DIao3
XOEeygJk_Go_QQGNVz6S44&hl=ar&sa=X&ei=KkmuUIqxHL
SM4gST04HYCg&ved=0CCkQ6AEwAg#v=onepage&q=%22t
he%20appellative%20function%20of%20phonic%20elem
ents%22&f=false
Surendra, D. Niyogi, P (2003) Measuring the Usefulness (Functional
Load) on Phonological Contrasts. Retrieved on Nov. 19th
2012 from:
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.5.203
Surendran and Levow, The functional load of tone in Mandarin is as
high as that of vowels, Proceedings of Speech Prosody 2004,
Nara, Japan, pp. 99-102.