In simplest terms, Morphophonology is the branch of linguistic theory which studies how allomorphs are phonologically represented. The question whether this particular subject should best be analysed as part of phonological or morphological theory seems in this respect irrelevant, as Morphophonology can be regarded as an interface between these two branches of linguistics and thus, regarding informative value, can be viewed equally well from either perspective (Kortmann 2005: 89). Later, I will address the issue of congruency, i.e. whether Morphophonology can be regarded as a separate sub-field of linguistics with independent theoretical properties and rules or rather as a merger between phonology and morphology. The latter point of view obviously begs the question whether morphological rules subsequently affect pronounciation or whether phonological properties determine not only the use of, but what is more, the need for allomorphs or in other words, whether, in terms of allomorphy, morphology precedes phonology or the other way around.
The aim of this paper is to present the different approaches to allomorphy as regards its phonological implications. After introducing the most important morphophonological rules of the English language, I will proceed to analysing the “Underlying Representations” on which these rules are based. Thus, following the theories first presented by Noam Chomsky and Morris Halle , I will focus on the psycholinguistic aspect of the URs as well as certain phenomena such as the so called “Trisyllabic Laxing” and sound changes or stress shifts conditioned by certain suffixes. The gigantic scope of Chomsky’s work on Universal Grammar makes it impossible to cover all of the aspects of URs as described in SPE, thus I will restrict my description of URs to only a few morphophonological phenomena that I find most fitting for this topic. Lastly, you will be presented with an alternating theory concerning the relation between allomorphy and phonology, namely Paul Kiparsky’s theory of Lexical Phonology , which postulates that an allomorph whose pronounciation in a certain morphological environment cannot be phonologically defined must be listed lexically (Spencer 1991: 118).
Contents
I. Introduction
II. Morphophonology
1. The Morphophonological Rules of the English Language
2. Underlying Representations
a) General Issues
b) The SPE Model
c) Suffix conditioning and Trisyllabic Laxing
3. Lexical Phonology
III. Conclusion
IV. Bibliography.
I. Introduction
In simplest terms, Morphophonology[1] is the branch of linguistic theory which studies how allomorphs are phonologically represented. The question whether this particular subject should best be analysed as part of phonological or morphological theory seems in this respect irrelevant, as Morphophonology can be regarded as an interface between these two branches of linguistics and thus, regarding informative value, can be viewed equally well from either perspective (Kortmann 2005: 89). Later, I will address the issue of congruency, i.e. whether Morphophonology can be regarded as a separate sub-field of linguistics with independent theoretical properties and rules or rather as a merger between phonology and morphology. The latter point of view obviously begs the question whether morphological rules subsequently affect pronounciation or whether phonological properties determine not only the use of, but what is more, the need for allomorphs or in other words, whether, in terms of allomorphy, morphology precedes phonology or the other way around.
The aim of this paper is to present the different approaches to allomorphy as regards its phonological implications. After introducing the most important morphophonological rules of the English language, I will proceed to analysing the “Underlying Representations”[2] on which these rules are based. Thus, following the theories first presented by Noam Chomsky and Morris Halle[3], I will focus on the psycholinguistic aspect of the URs as well as certain phenomena such as the so called “Trisyllabic Laxing” and sound changes or stress shifts conditioned by certain suffixes. The gigantic scope of Chomsky’s work on Universal Grammar makes it impossible to cover all of the aspects of URs as described in the SPE, thus I will restrict my description of URs to only a few morphophonological phenomena that I find most fitting for this topic. Lastly, you will be presented with an alternating theory concerning the relation between allomorphy and phonology, namely Paul Kiparsky’s theory of Lexical Phonology[4], which postulates that an allomorph whose pronounciation in a certain morphological environment can not be phonologically defined must be listed lexically (Spencer 1991: 118).
II. Morphophonology
1. The Morphophonological Rules of the English Language
In the English language we can find a number of rules which, based mainly on articulatory factors, determine the alternating pronunciation of morphemes. These morphemes, which differ in pronunciation depending on their phonological environment and in rare cases on special grammatical or lexical conditioning, are called allomorphs. Such phonologically conditioned allomorphs include the plural marker {-s} (/z/-/s/-/Iz/), the genitive and third person singular markers {-s}, which are phonologically identically presented, and the simple past and past participle marker {-d} (/d/-/t/-/Id/). Plural, genitive and third person singular markers are complementary distributed as follows:[5][6]
i) /z/ after all vowels and voiced consonants except sibilants /z/, /Z/, /Ù/, e.g. /bOIz/, /g3;lz/, /"lQr@z/, /ÙQnz/, /g@Uz/, /ri;dz/
ii) /s/ after all voiceless consonants except sibilants /s/, /S/, /Í/,e.g. /ru;fs/, /Ù&ks/, /kVts/
iii) /Iz/ after all sibilants, e.g. /"r@UzIz/, /"ru;ZIz/, /"ÙO;ÙIz/, /"k3;sIz/, /"rVSIz/, /"wIÍIz/
The complementary allomorph distribution for the simple past and past participle marker in regularly inflected verbs is as follows:
i) /d/ after all vowels and voiced consonants except /d/, e.g. /S@Ud/, /kO;ld/, /rQbd/,
ii) /t/ after all voiceless consonants except /t/, e.g. /pUSt/, /rIpt/, /pA;st/, /flInÍt/
iii) /Id/ after /d/ and /t/, e.g. /"feIdId/, /"l@UdId/, /"reItId/, /"pItId/
Unlike lexically conditioned plural allomorphs such as /gu;s/ - /gi;s/ (<goose>-<geese>) or past tense allomorphs such as /brIN/ - /brO;t/ (<bring>-<brought>), these allomorphs must not be learned by rote, but can rather be “predicted” depending on their phonological environment (Kortmann 2005: 90): a morpheme[7], e.g. the plural marker {-s} with the underlying phonological form /z/ goes through a process of phonological assimilation when combined with a morpheme that ends either with a sibilant or a voiceless consonant. This has simple articulatory reasons, i.e. our speech organs can not generate a voiced obstruent directly after a voiceless one and vice versa, making such phoneme pairs as */tz/ or */gs/ articulatory impossible without automatically creating another syllable e.g. /t@z/ or /g@s/.
This phonological conditioning based entirely on articulatory factors even allows us to create the (phonologically correct) plural forms of unknown or nonsense words such as plir, krote or floosh: /pl3;z/, /kr@Uts/ and /"flu;SIz/.[8]
Phonologically conditioned sound change does not only occur with inflected lexemes by the aforementioned grammatical affixes but also with derived lexemes by derivational affixes, i.e. certain prefixes adjust to the subsequent morpheme by differently pronounced allomorphs, whereas certain suffixes trigger one or more phonological changes in the preceding morpheme. One example for such a prefix morpheme is the negativity marker {in-} which occurs with a number of allomorphs:
i) /Im/ before bilabials and /m/, e.g. /Im"pQsIb@l/, /Im"b&l@ns/ or /I"mO;t@l/
ii) /IN/ before velars, e.g. /IN"kQns@kw@nt/ or /IN"glO;rI@s/[9]
iii) /Ir/ before /r/, e.g. /Ir@s"pQns@b@l/
iv) /In/ before all other consonants and all vowels, e.g. /Ins@"bO;dn@t/, /In"vQl@nt@rI/, /In"Qpr@b@l/, /In@"ksept@b@l/
Again, the sound change can be ascribed to an assimilation of the second (alternating) phoneme of the morpheme {in-}, in this case, a consonant, and the first consonant of the subsequent morpheme, creating the allomorphs /Im/, /IN/, /Ir/ and the “original” allomorph /In/, which is defined as the “Underlying Representation” of the morpheme (Katamba 1993: 32), a concept which will be further discussed in chapter 2.
Unlike prefixes and grammatical suffixes, derivational suffixes have no allomorphs at all, even more so, certain suffixes can change the phonological structure of the morpheme they are attached to, whether by a vowel or consonant change or a stress shift. An example for such a suffix would be the morpheme {-ity}, {state or quality of being X}, which can either lead to a vowel change ( /veIn/ + {-ity} à/"v&n@tI/), a stress shift ( /"rel@tIv/ + {-ity} à/rel@"tIv@tI/) or a consonant change and stress shift (/@"lektrik/ + {-ity} à/@lek"trIs@tI/).[10] This kind of suffix conditioning serves as an example for the morphological conditioning of allomorphs and will be further discussed in chapter 2.
In this paper I want to focus on rule-based phonological and morphological conditioning, so the grammatical allomorph conditioning of e.g. irregular verbs (such as /wi;p/ - /wept/) or the lexical allomorph conditioning of irregular plural forms (such as /fUt/ - /fi;t/) will not be discussed here.
[...]
[1] Although many linguists prefer the terms “Morphonology” or, especially in American linguistics, “Morphophonemics”, for purposes of clarity of the subject I will exclusively use the term “Morpho-phonology” throughout this paper.
[2] Further abbreviated as “UR” or (plural) “URs”.
[3] Chomsky, Noam and Morris Halle. 1968. The Sound Pattern of English. New York et. al.: Harper & Row. Further abbreviated as “SPE”.
[4] Further abbreviated as „LP“.
[5] Morphophonological rules are adopted from Kortmann 2005: 89-94 and Katamba 1993: 31-34.
[6] Please note that this distribution applies only for regular plural forms, thus lexically conditioned plural forms such as /Qks/ - /"Qks@n/ (<ox> - <oxen>) are not included in this rule.
[7] Note the distinction between the abstract term “morpheme” on the langue level and its concrete realisation on the parole level, the “allomorph” (Kortmann 2005: 90).
[8] Cf. Kortmann 2005: 90.
[9] Note that there is apparently a relatively new trend towards transcribing a velar nasal before a velar obstruent, e.g. <inglorious> was transcribed as /In"glO;ri@s/ 21 years ago. Cf. Messinger, Heinz (ed.). 21988. Langenscheidts Großes Schulwörterbuch Englisch-Deutsch. Nördlingen: C. H. Beck.
[10] Cf. Spencer 1991: 105 and Kortmann 2005: 92.
-
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X.