100% found this document useful (5 votes)
2K views

A Basic Introduction To Optimality Theory

Optimality Theory is a way of understanding how the human mind generates language. It was originally conceived as an alternative to Generative Phonology, but it has been extended to syntax and other domains of linguistics. This is a summary of Optimality Theory for introductory linguistics students, based in large part on Kager (1999) and incorporating other works as necessary.
Copyright
© Attribution (BY)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
100% found this document useful (5 votes)
2K views

A Basic Introduction To Optimality Theory

Optimality Theory is a way of understanding how the human mind generates language. It was originally conceived as an alternative to Generative Phonology, but it has been extended to syntax and other domains of linguistics. This is a summary of Optimality Theory for introductory linguistics students, based in large part on Kager (1999) and incorporating other works as necessary.
Copyright
© Attribution (BY)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

A Basic Introduction to Optimality Theory

Angus B. Grieve-Smith
Saint John's University
[email protected]

Optimality Theory is a way of understanding how the human mind generates


language. It was originally conceived as an alternative to Generative Phonology, but it
has been extended to syntax and other domains of linguistics. This is a summary of
Optimality Theory for introductory linguistics students, based in large part on Kager
(1999) and incorporating other works as necessary.

At the heart of Optimality Theory is the insight that some aspects of language
generation are easier to understand as a system of interacting constraints than as a
system of ordered rules. The constraints are universal, but each language variety puts
them in a unique ranking. The interactions among these constraints are represented with
tableaux; we will see some tableaux in the examples.

There are two kinds of constraints in Optimality. Markedness constraints include


the variety of ways that language users can make their languages easier to pronounce.
Faithfulness constraints assume that there is an underlying phonemic representation
(input) in the lexicon, and they prevent the words (output) from being distorted beyond
recognition by the markedness constraints. I have included a list of the most commonly
cited constraints at the end of this introduction, but here are ten that I will be focusing on
in the examples. The first five are markedness constraints, and the last five are
faithfulness constraints.

Name Function
Coda
*Voice Syllable codas are voiceless

*[] No []
*Labial No labial consonants
CV-Coord The consonant is identical to the onset of the following vowel gesture
CC-Coord The first of two consecutive consonants is identical to the second
Any segment in the input must have a corresponding segment in the
Dep-IO
output
Any segment in the input must have a corresponding segment in the
Max-IO
output
Ident-IO Segments have the same features in both input and output
Ident-IO(Place) Segments have the same place features in both input and output
Ident-IO(Voice) Segments have the same voicing features in both input and output
A Basic Introduction to Optimality Theory Angus B. Grieve-Smith

To illustrate the basics of Optimality, Kager (1999) analyzes the differences


between Dutch and English with respect to syllable-final voicing. In this respect, Dutch
behaves like German in the Language Files example.

Dutch:

1) /bd/ [bt] 'bed'


2) /bd-ən/ [bdən] 'beds'

English:

3) /bd/ [bd] 'bed'


4) /bd-ŋ/ [bdŋ] 'bedding'

In Kager's formulation, it comes down to the fact that Dutch ranks *VoiceCoda above
Ident-IO (Voice), while English ranks Ident-IO(Voice) higher. Kager uses the following
tableaux to show this:

Candidates *VoiceCoda Ident-IO (Voice)


a. ☞ [bt] *

b. [bd] *!

The constraint *VoiceCoda appears in the left column because it is higher-ranked


than Ident-IO (Voice). The * asterisk indicates a constraint violation. Because [bɛd]
violates the highest-ranked constraint, it is eliminated from consideration, as indicated by
the ! exclamation point, and the constraints that are not relevant to the final outcome are
shaded. [bɛt] violates the Ident-IO (Voice) constraint, but not the *VoiceCodaconstraint,
and it is therefore the optimal candidate, as indicated by the ☞ pointing hand. Here is a
similar tableau for English:

Candidates Ident-IO (Voice) *VoiceCoda


a. [bt] *!

b. ☞ [bd] *

We can borrow another example from the Language Files 4.2. In English,
syllable-final nasals assimilate to the place features of the following consonant, while in
French they do not:

2
A Basic Introduction to Optimality Theory Angus B. Grieve-Smith

5) I can ask -> [aj kæn æsk]


6) I can bake -> [aj kæm bek]
7) I can come -> [aj kæŋ km]
8) Donne-toi -> [dn twa]
9) Donne-moi -> [dn mwa]
10) Donne quoi? -> [dn kwa]

Optimality theory explains this place assimilation by ranking the basic Ident-IO
(Place) constraint below the CC-Coord constraint (Gafos, 2002). In order to allow
syllable-initial consonants to show distinctions of place, there is an Ident-IO-Onset
(Place) constraint that is ranked higher than these place markedness constraints.

/aj kæn bek/ Ident-IO-Onset (Place) CC-Coord Ident-IO (Place)


a. [aj kæn bek] *!

b. ☞ [aj kæm bek] *

c. [aj kæn dek] *!

For the contrasting case of French, the general Ident-IO (Place) constraint is
ranked higher than the place markedness constraints:

/dn
/dn mwa/ Ident-IO (Place) CC-Coord
a. ☞ [dn mwa] *

b. [dm mwa] *!

c. [dn nwa] *!

A similar analysis can be made for R-dropping in certain dialects (Orgun, 2001;
Abe, 2003). The Ident-IO-Onset (Place) constraint ensures that all onsets continue to
have the same place of articulation, while a *[] rule forbids [] from appearing anywhere
else in R-dropping dialects. In R-ful dialects, the *[] rule is ranked lower than Ident-IO-
Onset (Place):

3
A Basic Introduction to Optimality Theory Angus B. Grieve-Smith

/ð#/
/ð#/,
#/, /mæ#i/
/mæ#i/ (New York) Ident-IO-Onset (Place) *[#]
*[#] Ident-IO (Place)
a. [ð#] *!

b. ☞ [ðə] *

c. ☞ [mæ#i] *

d. [mæʔi] *! *

/ð#/
/ð#/,
#/, /mæ#i/
/mæ#i/ (Chicago) Ident-IO (Place) *[#]
*[#]
a. ☞ [ð#] *

b. [ðə] *!

c. ☞ [mæ#i] *

d. [mæʔi] *!

List of Constraints:
Many constraints have been proposed in Optimality Theory. There is a general
consensus that the theory is more useful the smaller the number of constraints, but that
doesn’t necessarily prevent people from proposing new ones. Here is a longer list of
constraints.

Name Function
Onset Syllables have onsets
*Coda Syllables have no codas
Onset
*Complex Syllable onsets are simple
*ComplexCoda Syllable codas are simple
*VoiceCoda Syllable codas are voiceless
Phonological entities (like words) begin and end at the same time as their
Align
corresponding grammatical entities

*[] No []

4
A Basic Introduction to Optimality Theory Angus B. Grieve-Smith

*[+low] No low vowels


*[+round] No rounded vowels
*[-back] No front vowels
*[+high] No high vowels
Cor-High Coronals (alveolar and dental consonants) are followed by high vowels
Coda-Cond
Place features are not distinctive in syllable codas
*Place
CV-Coord The consonant is identical to the onset of the following vowel gesture
CC-Coord The first consonant gesture is identical to the second
Dep-IO Any segment in the input must have a corresponding segment in the output
Max-IO Any segment in the input must have a corresponding segment in the output
Ident-IO Segments have the same features in both input and output
Ident-IO
Segments have the same place features in both input and output
(Place)
Ident-IO
Segments have the same manner features in both input and output
(Manner)
Ident-IO
Segments have the same voicing features in both input and output
(Voice)

References:
Abe, H. (2003). R dropping and R insertion in Received Pronunciation : A new look at an
old problem. Research reports of the Tsuruoka Technical College 38: 25-28.
Gafos, A. I. (2002). A grammar of gestural coordination. Natural Language and
Linguistic Theory 20: 269-337.
Kager, R. (1999). Optimality theory. Cambridge: Cambridge University Press.
Orgun, C. O. (2001). English r-insertion in Optimality Theory. Natural Language and
Linguistic Theory 19: 737-749.

You might also like