Scribd
Upload
32 views

Eskimo Data

This document analyzes the phonological rules governing vowel distribution in Greenlandic Eskimo. It identifies two rules: 1. [+high] front vowels [i] become [-high] front vowels [e] when followed by a consonant or #. 2. [+high] back vowels [u] become [-high] back vowels [o] when followed by a consonant or #. The analysis examines distribution patterns of vowels and identifies [i] and [u] as underlying phonemes with [e] and [o] as respective allophones in complementary distribution. It formulates feature-based rules to concisely describe the phonological processes.

Uploaded by

Kawtar Chadli
Copyright
© © All Rights Reserved
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
32 views

Eskimo Data

This document analyzes the phonological rules governing vowel distribution in Greenlandic Eskimo. It identifies two rules: 1. [+high] front vowels [i] become [-high] front vowels [e] when followed by a consonant or #. 2. [+high] back vowels [u] become [-high] back vowels [o] when followed by a consonant or #. The analysis examines distribution patterns of vowels and identifies [i] and [u] as underlying phonemes with [e] and [o] as respective allophones in complementary distribution. It formulates feature-based rules to concisely describe the phonological processes.

Uploaded by

Kawtar Chadli
Copyright
© © All Rights Reserved
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/ 3

LX 513 Phonology

Instructor: Prof. Trigo

First Assignment
Greenlandic Eskimo p. 43 (GP)
Jilani Warsi
ID # 611-16-8375

In stating a phonological rule, we need to ensure that the rule fulfills certain
requirements for it to be viable. A phonological rule should, therefore, subsume certain
properties that fulfill those requirements. A breakdown of those properties would be the
following:

X --> Y / A ___ B
where
a) Y has a distribution restriction, and is more predictable than X.
b) X does not have a distribution restriction, and is unpredictable.
c) Y is affected by either its preceding sound A or the following sound B.
d) Y is more marked than X, i.e., it is less frequent than X in many languages.
e) X is less marked than Y, i.e., its frequency of occurrence is relatively higher than Y.

Upon examining the list of both preceding and following segments of sounds in
Greenlandic Eskimo, the following linguistically relevant differences were found:

Phonetic Vowel Preceding Vowel Following Vowel

a n, r, m, v, s, q, l, g, i q, #, s, l, t, k, r
i #, s, q, t, k, p v, p, m, t, n, s, k, l, g, a
u l, t, n, k, # v, m, n, s, v, g
e p, n, s, m, n r, q, #
o n, l, #, k r, q, #

A closer scrutiny of the sounds preceding and following the Greenlandic phonetic
vowels reveals that there is a predictable distribution restriction on [e] and [o].
According to our requirement (a), both [e] and [o] should be Y, since they are always
followed by [r,q], or #, which is the place before the onset and after the coda. Assuming
[e] and [o] to be Y also fulfills our requirement (c). Thus, we will have the following
rule:

X --> [e, o]/_______ [consonant] or #


+ uvular
Now that we have determined what Y is, we should also determine what X may
be. Condition (b) states that X does not have distribution restriction, and is, therefore,
unpredictable. Since both [i] and [u] are never followed by [q,r], or # in Greenlandic, we
can assume that they are X. This is a valid assumption because these vowels have the
tendency to occur anywhere, which makes them unpredictable. Furthermore, X cannot
be [a] because as evident in our data, [a] is followed by [q,r], or #. This rules out the
possibility of [a] being X. On the basis of these facts, we will now rewrite our rule:

[ , u] --> [e, o] / _______ [consonant] or #

Our condition (d) requires that Y be more marked than X. A cross-linguistic


examination of the sound inventories of different languages would reveal that [ , u, a]
exist in many languages, whereas [e] and [o] do not. We stated earlier that the frequency
of occurrence is a relative determinant of markedness. Similarly, we can now safely
assume that [ , u,] are less marked than [e, o]. This fulfills our conditions (d) and (e).
In terms of vowel height, we see that condition (c) is an assimilation rule. That is,
in Greenlandic Eskimo the [+high] vowels [ , u] become [-high] vowels [e, o]. In this
particular case, our data shows that [+high] vowels assimilate to the context B, which is a
uvular consonant, or #. The assimilation of [+high] vowels to the following context will
yield the following rule:

[+high] --> [-high] / _____ [consonant] or #


vowel vowel +uvular

It is obvious that our rule operates in two different environments. It is, therefore,
advisable that we write two versions of the rule, but before we do that, we need to specify
one more feature. Since [ ] and [e] are both front vowels, it would be reasonable to
assume that [ ] becomes [e]. The same assumption can be made that [u] becomes [o]
because they are both back vowels. This assumption is in line with the ease of
articulation; the transition from front to front and back to back vowels is supposed to be
more natural and smooth. The transition from front to back and back to front vowels, on
the other hand, would be more cumbersome from an articulatory perspective. Therefore,
we will write the rules as:

(i) [+high] [-high]


[+front]--> [+front] / _________ [consonant] or #
[vowel] [vowel] +uvular

(ii) [+high] [-high]


[+back] -->[+back] / _________ [consonant] or #
[vowel] [vowel] +uvular

Since we stated that Y is more predictable because it manifests itself in the


environment in which it occurs, it would be reasonable to assume that according to our
data, y is an allophone of X. It only strengthens the claim that allophones are in
complementary distribution with their underlying phonemes. We can, therefore, state
that in Greenlandic Eskimo [ , u] are the two underlying vowels and [e, o] are their
respective allophones. This obviously leaves [a] as the third phoneme as it does not have
distribution restriction, and is unpredictable.

As mentioned previously, we can now write the two versions of each rule:

2
(ia) [+high] [-high]
[+front] --> [+front] / ________ [consonant]
[vowel] [vowel] +uvular

(ib) [+high] [-high]


[+front] --> [+front] / ________ #
[vowel] [vowel]

(iia) [+high] [-high]


[+back] --> [+back] / ________ [consonant]
[vowel] [vowel] +uvular

(iib) [+high] [-high]


[+back] --> [+back] / ________ #
[vowel] [vowel]

You might also like